refactor: isolate model families end to end by yifeif-nv · Pull Request #1093 · NVIDIA/TensorRT-Model-Connect

yifeif-nv · 2026-08-29T23:08:48Z

Background

The previous layout split each model across central builder, runtime,
registration, validation, dependency, and documentation systems. Adding or
changing one family therefore required shared model edits and increased
cross-family coupling.

This PR atomically replaces that layout with the architecture documented in
website/docs/architecture/ai-native-horizontal-scaling.md. It is an
intentional breaking cutover: no compatibility layer, migration path, legacy
fallback, or dual implementation remains.

Result

Every one of the 98 current-main families owns its complete vertical slice
under families/<family>/: build code, runtime DSO, dependency declaration,
manifests, thresholds, and direct tests.
Family production code does not import, include, link, inherit from, or
dispatch through sibling families.
Each model.py exposes one plain build(request, writer) function. Builder
inheritance is forbidden.
Native family pipelines implement the abstract Task APIs and use the thin
shared Engine/Bundle contracts.
core/builder/ is Python-only and core/runtime/ is C++-only. Shared
infrastructure contains no central family registry or optional model hooks.
Root build, packaging, CI selection, and documentation discover physical
family directories; normal family work stays inside its owner directory.

Preserved Functionality

Preserves BYOK as an explicit graph transform plus TVM-FFI runtime contract.
Preserves trtmc-bench, the native benchmark worker, dataset benchmark, and
release performance matrix as one-way consumers of public APIs.
Preserves the existing examples, including Cosmos3 dual-Spark,
FoundationPose, LeRobot ACT, VoiceChat full-duplex, and BYOK.
Preserves DevToolkit container lifecycle support as an application over the
public library boundary.
Preserves current MoGe behavior, K2-Horizon, Qwen3.8 FP8/NVFP4, and the five
timm families merged while this PR was open: DenseNet, EfficientNet,
Inception-v3, MNASNet, and RepVGG.
Restores the caller-provisioned SAM2 five-frame L4 qualification as an
explicit family-owned local command. It intentionally stays outside the
automated manifest inventory because its bundle, frames, and golden evidence
are not repository assets.

Deliberate Simplifications

Removes the old python/, src/, include/, tests/, benchmarks/, and
scripts/ architecture roots after moving active functionality to
core/, families/, apps/, and examples/.
Removes central model/runtime/workload/profile registries and generated
fallback reports.
Keeps one pinned container base image. Each family may declare its own plain
requirements.txt; there is no central dependency catalog.
Removes source, dependency, wheel, bundle, artifact, report, and profile
content-digest validation. Runtime checks retain only semantic format, shape,
ABI, ownership, and numerical correctness checks.
Accepts owner-local duplication to keep family dependencies one-way and
isolated.

Coverage Preservation

The dedicated matrix replays the protected behavior of the exact current-main
base and fails closed on missing or reassigned family/case ownership.

Current-main protected union: 98 families / 198 cases.
New pre-merge selection: 98 families / 104 cases.
New nightly selection: 98 families / 281 cases.
Required semantic assessment: 12 cases.
Missing protected families or cases: 0.

SAM2 L4 is additionally retained as a local-only qualification and is not
counted as an automated pass without caller-provisioned evidence.

Validation

Exact head: 9777fef79702c3d511a7e626617913d02101f45f

Public Community CPU exact-head workflow: all jobs passed.
- Source quality: passed, including 114 Python source/architecture tests.
- Docs: generated model inventory and production Docusaurus build passed.
- Ownership and impact: passed with the expected full-repository scope.
- Unit: 268 Python tests passed; fresh native build completed; CTest 65/65
  passed. Eight explicitly GPU-only CTests were skipped in the hardened CPU
  lane rather than counted as passes.
Architecture isolation suite: 38 passed.
tools/model_ci.py validate: valid for all 98 families.
Production package lane:
- sdist includes every family dependency declaration;
- wheel includes all family Python/data and exactly 98 family DSOs;
- installed trtmc, trtmc-bench, and bundle inspection passed;
- every installed DSO loaded with RTLD_NOW.
Focused five-timm and SAM2 L4 port suites: 64 Python tests passed, two
expected unselected-GPU skips; five timm DSOs and five preprocessing tests
built and passed in both focused and full native builds.
SAM2 L4 local runner device-free contracts passed; malformed, missing,
symlinked, incomplete, and accuracy-regressed inputs fail closed.
Ruff, clang-format dry run, legal-header audit, complexity gate, package
checks, and git diff --check: passed.
DCO and PR metadata checks: passed.

Remaining Gate

The dedicated protected pre-merge and nightly lanes must pass on this exact
head before merge. They provide real checkpoint/native/reference coverage that
is intentionally unavailable in the public CPU lane. No merge is requested by
this update.

Review Notes

Suggested order: architecture document; thin core contracts; two
structurally different family slices; Task/Engine interfaces; apps/examples;
closed-set architecture tests; CI/package selection.
Do not extract duplicated family code into shared model helpers without a
concrete stable cross-family contract.
Existing bundles must be rebuilt against the new public contracts.

Change categories

Risk level

Low
Medium
High

This deliberately replaces every family build/runtime/test path, packaging
layout, and core API/ABI boundary in one incompatible change.

coderabbitai · 2026-08-29T23:09:06Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-08-29T23:29:27Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-08-29T23:49:27Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-08-30T00:09:17Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-08-30T00:36:19Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-08-30T00:56:20Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-08-30T01:44:51Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-08-30T07:05:07Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-08-30T09:38:21Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-08-30T09:55:22Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-08-31T17:15:54Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-08-31T17:57:37Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-09-01T18:41:38Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-09-01T20:21:50Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-09-01T20:25:10Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

github-actions · 2026-09-01T21:24:20Z

This is an automated Internal CI result; no review from an individual maintainer is requested.

TRTMC Protected CI failure
==========================

This log contains only approved structured failure fields.

Status: FAILED
Repository: NVIDIA/TensorRT-Model-Connect
Pull request: #1093
Head commit: 82818497469bc7ce69a95e5467f12c40e4e4c0d2
Tested revision: 335b931f366e55f186265c8650d31f1a2b05e899 (merge)
Run attempt: 1
Generated at: 2026-09-01T21:23:44Z
Disclosure policy: 2026-08-27

Failure summary
---------------
Failure 1
  Class: legal_error
  Reason: canonical_document_mismatch
  Cause: A canonical repository document does not match policy.
  Stage: legal
  Model: other-model
  Backend: other-backend
  GPU: protected-gpu
  Test: CONTRIBUTING.md

Report ID: trtmc-pr1093-8281849-attempt1

Open the public Source Actions run from the automated status link above.

coderabbitai · 2026-09-01T21:27:01Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-09-02T02:45:48Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-09-02T03:03:22Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-09-02T03:24:41Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-09-02T03:40:57Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-09-02T03:59:50Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-09-02T04:02:07Z

Caution

CodeRabbit couldn't post its review summary.

Error details

Validation Failed: {"resource":"IssueComment","code":"unprocessable","field":"data","message":"Body is too long (maximum is 65536 characters)"} - https://docs.github.com/rest/issues/comments#create-an-issue-comment

coderabbitai · 2026-09-02T04:47:10Z

Important

Review skipped

Too many files!

This PR contains 2966 files, which is 2666 over the limit of 300.

To get a review, reduce the PR to 300 files or fewer by splitting it into smaller PRs or changing its base branch.

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 0426ae7b-adfc-4f98-af46-f187f7a036f5

📥 Commits

Reviewing files that changed from the base of the PR and between 92db111 and 9777fef.

⛔ Files ignored due to path filters (34)

families/canary/tests/data/Recording.wav is excluded by !**/*.wav
families/canary/tests/data/asr_probes/probe_01_clean_48k_stereo_baseline.wav is excluded by !**/*.wav
families/canary/tests/data/asr_probes/probe_02_clean_16k_mono_no_resample.wav is excluded by !**/*.wav
families/canary/tests/data/asr_probes/probe_03_clean_48k_mono_resample.wav is excluded by !**/*.wav
families/canary/tests/data/asr_probes/probe_04_clean_48k_stereo_gain_skew.wav is excluded by !**/*.wav
families/canary/tests/data/asr_probes/probe_05_low_volume_48k_stereo.wav is excluded by !**/*.wav
families/canary/tests/data/asr_probes/probe_06_leading_trailing_silence_48k_stereo.wav is excluded by !**/*.wav
families/canary/tests/data/asr_probes/probe_08_noisy_48k_stereo_snr20.wav is excluded by !**/*.wav
families/deepseek_ocr/tests/data/orc_test_img.jpeg is excluded by !**/*.jpeg
families/dinov3/tests/data/test_img.jpeg is excluded by !**/*.jpeg
families/fast_foundation_stereo/tests/data/office_left.png is excluded by !**/*.png
families/fast_foundation_stereo/tests/data/office_right.png is excluded by !**/*.png
families/internvl/tests/data/test_img.jpeg is excluded by !**/*.jpeg
families/lance/tests/data/test_img.jpeg is excluded by !**/*.jpeg
families/lerobot_act/tests/data/recorded_observation/observation.images.top.png is excluded by !**/*.png
families/locateanything/tests/data/test_img.jpeg is excluded by !**/*.jpeg
families/moge/tests/data/test_img.jpeg is excluded by !**/*.jpeg
families/nemotron_speech_streaming/tests/data/Recording.wav is excluded by !**/*.wav
families/nemotron_speech_streaming/tests/data/asr_probes/probe_01_clean_48k_stereo_baseline.wav is excluded by !**/*.wav
families/nemotron_speech_streaming/tests/data/asr_probes/probe_02_clean_16k_mono_no_resample.wav is excluded by !**/*.wav
families/nemotron_speech_streaming/tests/data/asr_probes/probe_03_clean_48k_mono_resample.wav is excluded by !**/*.wav
families/nemotron_speech_streaming/tests/data/asr_probes/probe_04_clean_48k_stereo_gain_skew.wav is excluded by !**/*.wav
families/nemotron_speech_streaming/tests/data/asr_probes/probe_05_low_volume_48k_stereo.wav is excluded by !**/*.wav
families/nemotron_speech_streaming/tests/data/asr_probes/probe_06_leading_trailing_silence_48k_stereo.wav is excluded by !**/*.wav
families/nemotron_speech_streaming/tests/data/asr_probes/probe_08_noisy_48k_stereo_snr20.wav is excluded by !**/*.wav
families/nemotron_voicechat/tests/assets/sample_general_input.flac is excluded by !**/*.flac
families/nemotron_voicechat/tests/assets/sample_general_reference.flac is excluded by !**/*.flac
families/personaplex/tests/data/Recording.wav is excluded by !**/*.wav
families/phi4_multimodal/tests/data/test_img.jpeg is excluded by !**/*.jpeg
families/qwen_image/tests/data/test_img.jpeg is excluded by !**/*.jpeg
families/qwen_vl/tests/data/test_img.jpeg is excluded by !**/*.jpeg
families/sam/tests/data/test_img.jpeg is excluded by !**/*.jpeg
families/sam3/tests/data/test_img.jpeg is excluded by !**/*.jpeg
families/sana_wm/tests/assets/demo_0.png is excluded by !**/*.png

📒 Files selected for processing (2966)

.coderabbit.yaml
.dockerignore
.github/scripts/build-python-profiles.py
.github/scripts/write-model-proof-fallback-report.py
.github/workflows/community-cpu.yml
.github/workflows/internal-ci-bridge.yml
.github/workflows/pages.yml
.gitignore
AGENTS.md
ASSET_LICENSES.md
CMakeLists.txt
CONTRIBUTING.md
Dockerfile
Dockerfile.community-cpu
Dockerfile.dev.aarch64
Dockerfile.dev.x86
README.md
_pyproject_backend.py
apps/benchmark/example.yaml
apps/benchmark/native/benchmark_worker.cpp
apps/benchmark/native/dataset_answer.h
apps/benchmark/native/dataset_benchmark.cpp
apps/benchmark/performance/README.md
apps/benchmark/performance/baselines/audio_reference.py
apps/benchmark/performance/baselines/elf_reference.py
apps/benchmark/performance/baselines/hf_transformers.py
apps/benchmark/performance/baselines/lance_reference.py
apps/benchmark/performance/baselines/reference_support.py
apps/benchmark/performance/baselines/sana_wm_reference.py
apps/benchmark/performance/baselines/task_reference.py
apps/benchmark/performance/baselines/timing_contracts.py
apps/benchmark/performance/environments/auto-thor.yaml
apps/benchmark/performance/environments/gb300.yaml
apps/benchmark/performance/environments/l4t-thor.yaml
apps/benchmark/performance/release.yaml
apps/benchmark/performance/requirements.txt
apps/benchmark/tests/native/test_benchmark_worker_e2e.cpp
apps/benchmark/tests/native/test_dataset_answer.cpp
apps/benchmark/trtmc-bench
apps/benchmark/trtmc_benchmark/__init__.py
apps/benchmark/trtmc_benchmark/__main__.py
apps/benchmark/trtmc_benchmark/builder.py
apps/benchmark/trtmc_benchmark/catalog.py
apps/benchmark/trtmc_benchmark/cli.py
apps/benchmark/trtmc_benchmark/metrics.py
apps/benchmark/trtmc_benchmark/operations.py
apps/benchmark/trtmc_benchmark/report.py
apps/benchmark/trtmc_benchmark/service.py
apps/benchmark/trtmc_benchmark/task_adapters.py
apps/benchmark/trtmc_benchmark/telemetry.py
apps/benchmark/trtmc_benchmark/tests/test_benchmark.py
apps/benchmark/trtmc_benchmark/tests/test_perf_matrix.py
apps/benchmark/trtmc_benchmark/types.py
apps/benchmark/trtmc_benchmark/worker.py
apps/cli/cli.cpp
apps/cli/cli.h
apps/cli/io.cpp
apps/cli/io.h
apps/cli/main.cpp
apps/cli/tests/test_cli.cpp
apps/devtoolkit/README.md
apps/devtoolkit/examples/prepare_environment.py
apps/devtoolkit/trtmc_devtoolkit/__init__.py
apps/devtoolkit/trtmc_devtoolkit/api.py
apps/devtoolkit/trtmc_devtoolkit/docker_target.py
apps/devtoolkit/trtmc_devtoolkit/models.py
apps/devtoolkit/trtmc_devtoolkit/runner.py
benchmarks/performance/README.md
benchmarks/performance/baselines/hf_transformers.py
benchmarks/performance/baselines/task_reference.py
benchmarks/performance/environments/auto-thor.yaml
benchmarks/performance/environments/gb300.yaml
benchmarks/performance/environments/l4t-thor.yaml
cmake/model_plugin_index.cpp.in
cmake/register_model_plugin.cpp.in
cmake/register_schemas.cpp.in
cmake/trtmcConfig.cmake.in
cmake/trtmc_config_schemas.cmake
cmake/trtmc_pipeline_plugins.cmake
cmake/trtmc_registration_manifest.cmake
conanfile.py
conftest.py
core/builder/tensorrt_model_connect/__init__.py
core/builder/tensorrt_model_connect/__main__.py
core/builder/tensorrt_model_connect/build.py
core/builder/tensorrt_model_connect/build_cli.py
core/builder/tensorrt_model_connect/bundle_writer.py
core/builder/tensorrt_model_connect/byok.py
core/builder/tensorrt_model_connect/graph_transform.py
core/builder/tensorrt_model_connect/model_support.py
core/builder/tests/__init__.py
core/builder/tests/test_build.py
core/builder/tests/test_build_cli.py
core/builder/tests/test_bundle_writer.py
core/builder/tests/test_byok.py
core/builder/tests/test_graph_transform.py
core/builder/tests/test_model_support.py
core/runtime/bundle/bundle_format.cpp
core/runtime/bundle/bundle_format.h
core/runtime/byok/byok.cpp
core/runtime/byok/tvm_ffi_function.cpp
core/runtime/byok/tvm_ffi_function.h
core/runtime/byok/tvm_ffi_kernel_creator.cpp
core/runtime/byok/tvm_ffi_kernel_plugin.cpp
core/runtime/byok/tvm_ffi_kernel_plugin.h
core/runtime/include/trtmc/bundle.h
core/runtime/include/trtmc/byok.h
core/runtime/include/trtmc/runtime/device_tensor.h
core/runtime/include/trtmc/runtime/family_factory.h
core/runtime/include/trtmc/runtime/family_loader.h
core/runtime/include/trtmc/runtime/tensor.h
core/runtime/include/trtmc/runtime/trt_backend.h
core/runtime/include/trtmc/runtime/trt_module.h
core/runtime/include/trtmc/task.h
core/runtime/loader/family_loader.cpp
core/runtime/primitives/cuda_common.cpp
core/runtime/primitives/cuda_common.h
core/runtime/primitives/device_tensor.cpp
core/runtime/primitives/trt_common.cpp
core/runtime/primitives/trt_common.h
core/runtime/tensorrt/trt_backend.cpp
core/runtime/tensorrt/trt_logger.cpp
core/runtime/tensorrt/trt_logger.h
core/runtime/tensorrt/trt_module_impl.cpp
core/runtime/tensorrt/trt_module_impl.h
core/runtime/tests/fake_backend.cpp
core/runtime/tests/fake_family.cpp
core/runtime/tests/test_bundle_format_v1.cpp
core/runtime/tests/test_byok_shape_spec.cpp
core/runtime/tests/test_family_loader.cpp
core/runtime/tests/test_task_api.cpp
examples/byok/README.md
examples/byok/export_cutedsl_residual_add.py
examples/byok/test_tvm_ffi_plugin.cpp
examples/models/cosmos3/dual_spark/Dockerfile
examples/models/cosmos3/dual_spark/Dockerfile.dockerignore
examples/models/cosmos3/dual_spark/README.md
examples/models/cosmos3/dual_spark/run_dual_spark.py
examples/models/cosmos3/dual_spark/test_cosmos3_dual_spark_source.py
examples/models/foundationpose/preprocessed_refinement/CMakeLists.txt
examples/models/foundationpose/preprocessed_refinement/README.md
examples/models/foundationpose/preprocessed_refinement/main.cpp
examples/models/foundationpose/preprocessed_refinement/qualification/gb300-trt11.1-fp16.json
examples/models/foundationpose/preprocessed_refinement/qualification/gb300-trt11.1-fp32.json
examples/models/lerobot_act/recorded_control/CMakeLists.txt
examples/models/lerobot_act/recorded_control/README.md
examples/models/lerobot_act/recorded_control/main.cpp
examples/models/lerobot_act/recorded_control/qualification/gb300-trt11.1-fp32.json
examples/models/nemotron_voicechat/full_duplex/CMakeLists.txt
examples/models/nemotron_voicechat/full_duplex/Dockerfile
examples/models/nemotron_voicechat/full_duplex/Dockerfile.dockerignore
examples/models/nemotron_voicechat/full_duplex/README.md
examples/models/nemotron_voicechat/full_duplex/main.cpp
examples/models/nemotron_voicechat/full_duplex/test_voicechat_full_duplex_source.py
examples/trtmc_bench.yaml
examples/trtmc_benchmark_worker.cpp
examples/trtmc_dataset_benchmark.cpp
families/__init__.py
families/albert/__init__.py
families/albert/checkpoint_mapper.py
families/albert/config.py
families/albert/encoder_builder.py
families/albert/graph_ops.py
families/albert/model.py
families/albert/parallel.py
families/albert/runtime/CMakeLists.txt
families/albert/runtime/pipeline.cpp
families/albert/runtime/pipeline.h
families/albert/runtime/plugin.cpp
families/albert/runtime/plugin_helpers.cpp
families/albert/runtime/plugin_helpers.h
families/albert/runtime/tokenizer.h
families/albert/runtime/unigram_tokenizer.cpp
families/albert/support.py
families/albert/tests/__init__.py
families/albert/tests/manifests/albert-base-tp4.json
families/albert/tests/manifests/albert-base.json
families/albert/tests/test_e2e.py
families/albert/tests/thresholds/albert-base-tp4.json
families/albert/tests/thresholds/albert-base.json
families/albert/tp_builder.py
families/bark/__init__.py
families/bark/checkpoint_mapper.py
families/bark/config.py
families/bark/decoder_tp_builder.py
families/bark/default_decoder.py
families/bark/default_dual_profile_decoder.py
families/bark/encodec_builder.py
families/bark/graph_blocks.py
families/bark/graph_ops.py
families/bark/model.py
families/bark/parallel.py
families/bark/requirements.txt
families/bark/runtime/CMakeLists.txt
families/bark/runtime/bark_config.h
families/bark/runtime/bark_generation_plan.h
families/bark/runtime/decode_runtime.cpp
families/bark/runtime/decode_runtime.h
families/bark/runtime/inference_state.h
families/bark/runtime/kv_cache.cpp
families/bark/runtime/kv_cache.h
families/bark/runtime/pipeline.cpp
families/bark/runtime/pipeline.h
families/bark/runtime/plugin.cpp
families/bark/runtime/plugin_helpers.cpp
families/bark/runtime/plugin_helpers.h
families/bark/runtime/sampler.cpp
families/bark/runtime/sampler.h
families/bark/runtime/sparse_multinomial_kernel.cu
families/bark/runtime/sparse_multinomial_kernel.h
families/bark/runtime/tokenizer.h
families/bark/runtime/wordpiece_tokenizer.cpp
families/bark/support.py
families/bark/tests/__init__.py
families/bark/tests/cpp/bark_token_trace.cpp
families/bark/tests/manifests/bark-large-tp4.json
families/bark/tests/manifests/bark-large.json
families/bark/tests/manifests/bark-small-fp32-l0.json
families/bark/tests/manifests/bark-small-tp4.json
families/bark/tests/manifests/bark-small.json
families/bark/tests/test_e2e.py
families/bark/tests/thresholds/bark-large-tp4.json
families/bark/tests/thresholds/bark-large.json
families/bark/tests/thresholds/bark-small-fp32-l0.json
families/bark/tests/thresholds/bark-small-tp4.json
families/bark/tests/thresholds/bark-small-tts-probe01.json
families/bark/tests/thresholds/bark-small-tts-probe02.json
families/bark/tests/thresholds/bark-small.json
families/bark/utils.py
families/bart/__init__.py
families/bart/checkpoint_mapper.py
families/bart/config.py
families/bart/decoder_tp_builder.py
families/bart/graph_ops.py
families/bart/model.py
families/bart/parallel.py
families/bart/runtime/CMakeLists.txt
families/bart/runtime/bpe_tokenizer.cpp
families/bart/runtime/decode_runtime.cpp
families/bart/runtime/decode_runtime.h
families/bart/runtime/distributed_runtime.cpp
families/bart/runtime/distributed_runtime.h
families/bart/runtime/inference_state.h
families/bart/runtime/kv_cache.cpp
families/bart/runtime/kv_cache.h
families/bart/runtime/plugin.cpp
families/bart/runtime/plugin_helpers.cpp
families/bart/runtime/plugin_helpers.h
families/bart/runtime/runtime_config.h
families/bart/runtime/tokenizer.h
families/bart/support.py
families/bart/tests/__init__.py
families/bart/tests/cpp/test_bart_runtime_config.cpp
families/bart/tests/manifests/bart-base-tp4.json
families/bart/tests/manifests/bart-base.json
families/bart/tests/test_e2e.py
families/bart/tests/thresholds/bart-base-tp4.json
families/bart/tests/thresholds/bart-base.json
families/bert/__init__.py
families/bert/config.py
families/bert/model.py
families/bert/parallel.py
families/bert/runtime/CMakeLists.txt
families/bert/runtime/pipeline.cpp
families/bert/runtime/pipeline.h
families/bert/runtime/plugin.cpp
families/bert/runtime/plugin_helpers.cpp
families/bert/runtime/plugin_helpers.h
families/bert/runtime/tokenizer.h
families/bert/runtime/unigram_tokenizer.cpp
families/bert/runtime/wordpiece_tokenizer.cpp
families/bert/support.py
families/bert/tests/__init__.py
families/bert/tests/manifests/all-minilm-l6-v2.json
families/bert/tests/manifests/bert-base-uncased-tp4.json
families/bert/tests/manifests/bert-base-uncased.json
families/bert/tests/manifests/bge-small-en-v1.5.json
families/bert/tests/manifests/multilingual-e5-small.json
families/bert/tests/manifests/paraphrase-multilingual-minilm-l12-v2.json
families/bert/tests/test_e2e.py
families/bert/tests/test_weights.py
families/bert/tests/thresholds/all-minilm-l6-v2.json
families/bert/tests/thresholds/bert-base-uncased-tp4.json
families/bert/tests/thresholds/bert-base-uncased.json
families/bert/tests/thresholds/bge-small-en-v1.5.json
families/bert/tests/thresholds/multilingual-e5-small-passage-zh.json
families/bert/tests/thresholds/multilingual-e5-small-query-en.json
families/bert/tests/thresholds/paraphrase-multilingual-minilm-l12-v2.json
families/bert/weights/__init__.py
families/bloom/__init__.py
families/bloom/checkpoint_mapper.py
families/bloom/config.py
families/bloom/dual_profile_decoder_builder.py
families/bloom/dual_profile_decoder_tp_builder.py
families/bloom/graph_blocks.py
families/bloom/graph_ops.py
families/bloom/model.py
families/bloom/parallel.py
families/bloom/runtime/CMakeLists.txt
families/bloom/runtime/bpe_tokenizer.cpp
families/bloom/runtime/chat_templates.cpp
families/bloom/runtime/chat_templates.h
families/bloom/runtime/distributed_runtime.cpp
families/bloom/runtime/distributed_runtime.h
families/bloom/runtime/inference_state.h
families/bloom/runtime/kv_cache.cpp
families/bloom/runtime/kv_cache.h
families/bloom/runtime/pipeline.cpp
families/bloom/runtime/pipeline.h
families/bloom/runtime/plugin.cpp
families/bloom/runtime/plugin_helpers.cpp
families/bloom/runtime/plugin_helpers.h
families/bloom/runtime/sampler.cpp
families/bloom/runtime/sampler.h
families/bloom/runtime/tensor_names.h
families/bloom/runtime/tokenizer.h
families/bloom/standard_decoder_builder.py
families/bloom/support.py
families/bloom/tests/__init__.py
families/bloom/tests/manifests/bloom-560m-tp4.json
families/bloom/tests/manifests/bloom-560m.json
families/bloom/tests/test_e2e.py
families/bloom/tests/thresholds/bloom-560m-tp4.json
families/bloom/tests/thresholds/bloom-560m.json
families/canary/__init__.py
families/canary/batching.py
families/canary/checkpoint_mapper.py
families/canary/config.py
families/canary/decoder_tp_builder.py
families/canary/graph_blocks.py
families/canary/graph_ops.py
families/canary/model.py
families/canary/parallel.py
families/canary/requirements.txt
families/canary/runtime/CMakeLists.txt
families/canary/runtime/canary_config.h
families/canary/runtime/canary_cross_kv_apply.h
families/canary/runtime/canary_cross_kv_plan.h
families/canary/runtime/canary_decode_policy.h
families/canary/runtime/canary_host_plan.h
families/canary/runtime/canary_mel_spectrogram.cpp
families/canary/runtime/canary_mel_spectrogram.h
families/canary/runtime/canary_request.h
families/canary/runtime/canary_segment_utils.h
families/canary/runtime/decode_runtime.cpp
families/canary/runtime/decode_runtime.h
families/canary/runtime/inference_state.h
families/canary/runtime/kv_cache.cpp
families/canary/runtime/kv_cache.h
families/canary/runtime/pipeline.cpp
families/canary/runtime/pipeline.h
families/canary/runtime/plugin.cpp
families/canary/runtime/plugin_helpers.cpp
families/canary/runtime/plugin_helpers.h
families/canary/runtime/resampler.cpp
families/canary/runtime/resampler.h
families/canary/runtime/tokenizer.h
families/canary/runtime/unigram_tokenizer.cpp
families/canary/support.py
families/canary/tests/__init__.py
families/canary/tests/data/asr_probes/README.md
families/canary/tests/data/asr_probes/generate_asr_probe_inputs.py
families/canary/tests/data/asr_probes/manifest.json
families/canary/tests/manifests/canary-1b-v2-tp4.json
families/canary/tests/manifests/canary-1b-v2.json
families/canary/tests/test_e2e.py
families/canary/tests/thresholds/canary-1b-v2-asr-probe01.json
families/canary/tests/thresholds/canary-1b-v2-asr-probe02.json
families/canary/tests/thresholds/canary-1b-v2-asr-probe03.json
families/canary/tests/thresholds/canary-1b-v2-asr-probe04.json
families/canary/tests/thresholds/canary-1b-v2-asr-probe05.json
families/canary/tests/thresholds/canary-1b-v2-asr-probe06.json
families/canary/tests/thresholds/canary-1b-v2-asr-probe08.json
families/canary/tests/thresholds/canary-1b-v2-tp4.json
families/canary/tests/thresholds/canary-1b-v2.json
families/chronos_bolt/__init__.py
families/chronos_bolt/checkpoint_mapper.py
families/chronos_bolt/config.py
families/chronos_bolt/graph_ops.py
families/chronos_bolt/model.py
families/chronos_bolt/requirements.txt
families/chronos_bolt/runtime/CMakeLists.txt
families/chronos_bolt/runtime/pipeline.cpp
families/chronos_bolt/runtime/pipeline.h
families/chronos_bolt/runtime/plugin.cpp
families/chronos_bolt/support.py
families/chronos_bolt/tests/__init__.py
families/chronos_bolt/tests/cpp/test_forecast_contract.cpp
families/chronos_bolt/tests/etth1.py
families/chronos_bolt/tests/manifests/chronos-bolt-tiny-official-tp4.json
families/chronos_bolt/tests/manifests/chronos-bolt-tiny-official.json
families/chronos_bolt/tests/reference-source.json
families/chronos_bolt/tests/test_e2e.py
families/chronos_bolt/tests/test_etth1.py
families/chronos_bolt/tests/thresholds/chronos-bolt-tiny-official-tp4.json
families/chronos_bolt/tests/thresholds/chronos-bolt-tiny-official.json
families/chronos_bolt/time_series_trt.py
families/codegen/__init__.py
families/codegen/checkpoint_mapper.py
families/codegen/config.py
families/codegen/default_decoder.py
families/codegen/default_dual_profile_decoder.py
families/codegen/dual_profile_decoder_tp_builder.py
families/codegen/graph_blocks.py
families/codegen/graph_ops.py
families/codegen/model.py
families/codegen/parallel.py
families/codegen/runtime/CMakeLists.txt
families/codegen/runtime/bpe_tokenizer.cpp
families/codegen/runtime/chat_templates.cpp
families/codegen/runtime/chat_templates.h
families/codegen/runtime/distributed_runtime.cpp
families/codegen/runtime/distributed_runtime.h
families/codegen/runtime/inference_state.h
families/codegen/runtime/kv_cache.cpp
families/codegen/runtime/kv_cache.h
families/codegen/runtime/pipeline.cpp
families/codegen/runtime/pipeline.h
families/codegen/runtime/plugin.cpp
families/codegen/runtime/plugin_helpers.cpp
families/codegen/runtime/plugin_helpers.h
families/codegen/runtime/sampler.cpp
families/codegen/runtime/sampler.h
families/codegen/runtime/tensor_names.h
families/codegen/runtime/tokenizer.h
families/codegen/support.py
families/codegen/tests/__init__.py
families/codegen/tests/manifests/codegen-350m-tp4.json
families/codegen/tests/manifests/codegen-350m.json
families/codegen/tests/test_e2e.py
families/codegen/tests/thresholds/codegen-350m-tp4.json
families/codegen/tests/thresholds/codegen-350m.json
families/codegen/utils.py
families/convbert/__init__.py
families/convbert/builder.py
families/convbert/checkpoint_mapper.py
families/convbert/config.py
families/convbert/graph_ops.py
families/convbert/model.py
families/convbert/parallel.py
families/convbert/runtime/CMakeLists.txt
families/convbert/runtime/pipeline.cpp
families/convbert/runtime/pipeline.h
families/convbert/runtime/plugin.cpp
families/convbert/runtime/plugin_helpers.cpp
families/convbert/runtime/plugin_helpers.h
families/convbert/runtime/tokenizer.h
families/convbert/runtime/wordpiece_tokenizer.cpp
families/convbert/support.py
families/convbert/tests/__init__.py
families/convbert/tests/manifests/convbert-base-tp2.json
families/convbert/tests/manifests/convbert-base.json
families/convbert/tests/test_e2e.py
families/convbert/tests/thresholds/convbert-base-tp2.json
families/convbert/tests/thresholds/convbert-base.json
families/convbert/tp_builder.py
families/cosmos3/__init__.py
families/cosmos3/checkpoint_mapper.py
families/cosmos3/graph_blocks.py
families/cosmos3/graph_ops.py
families/cosmos3/model.py
families/cosmos3/model_config.py
families/cosmos3/requirements.txt
families/cosmos3/runtime/CMakeLists.txt
families/cosmos3/runtime/bpe_tokenizer.cpp
families/cosmos3/runtime/conditioning.cpp
families/cosmos3/runtime/conditioning.h
families/cosmos3/runtime/cosmos3_unipc_coefficients.h
families/cosmos3/runtime/cosmos3_unipc_cuda.cu
families/cosmos3/runtime/cosmos3_unipc_cuda.h
families/cosmos3/runtime/distributed_runtime.cpp
families/cosmos3/runtime/distributed_runtime.h
families/cosmos3/runtime/pipeline.cpp
families/cosmos3/runtime/pipeline.h
families/cosmos3/runtime/plugin.cpp
families/cosmos3/runtime/runtime_config.cpp
families/cosmos3/runtime/runtime_config.h
families/cosmos3/runtime/tokenizer.h
families/cosmos3/runtime/torch_cuda_normal.cu
families/cosmos3/runtime/torch_cuda_normal.h
families/cosmos3/runtime/vae_cache_storage.cpp
families/cosmos3/runtime/vae_cache_storage.h
families/cosmos3/support.py
families/cosmos3/tests/__init__.py
families/cosmos3/tests/cpp/test_cosmos3_conditioning.cpp
families/cosmos3/tests/cpp/test_cosmos3_runtime_config.cpp
families/cosmos3/tests/cpp/test_cosmos3_vae_cache.cpp
families/cosmos3/tests/manifests/cosmos3-nano-l0-cp2.json
families/cosmos3/tests/manifests/cosmos3-nano-l0.json
families/cosmos3/tests/test_e2e.py
families/cosmos3/tests/test_model.py
families/cosmos3/tests/thresholds/cosmos3-nano-l0-cp2.json
families/cosmos3/tests/thresholds/cosmos3-nano-l0.json
families/cosmos3/transformer_builder.py
families/cosmos3/trt_ops.py
families/cosmos3/vae_step_builder.py
families/deberta/__init__.py
families/deberta/config.py
families/deberta/graph/__init__.py
families/deberta/graph/model.py
families/deberta/model.py
families/deberta/parallel.py
families/deberta/runtime/CMakeLists.txt
families/deberta/runtime/bpe_tokenizer.cpp
families/deberta/runtime/pipeline.cpp
families/deberta/runtime/pipeline.h
families/deberta/runtime/plugin.cpp
families/deberta/runtime/plugin_helpers.cpp
families/deberta/runtime/plugin_helpers.h
families/deberta/runtime/tokenizer.h
families/deberta/support.py
families/deberta/tests/__init__.py
families/deberta/tests/manifests/deberta-base-tp4.json
families/deberta/tests/manifests/deberta-base.json
families/deberta/tests/test_e2e.py
families/deberta/tests/thresholds/deberta-base-tp4.json
families/deberta/tests/thresholds/deberta-base.json
families/deberta/weights/__init__.py
families/deepseek_ocr/__init__.py
families/deepseek_ocr/checkpoint_mapper.py
families/deepseek_ocr/config.py
families/deepseek_ocr/default_decoder.py
families/deepseek_ocr/default_dual_profile_decoder.py
families/deepseek_ocr/graph_blocks.py
families/deepseek_ocr/graph_ops.py
families/deepseek_ocr/model.py
families/deepseek_ocr/parallel.py
families/deepseek_ocr/prefill_config.py
families/deepseek_ocr/requirements.txt
families/deepseek_ocr/runtime/CMakeLists.txt
families/deepseek_ocr/runtime/bpe_tokenizer.cpp
families/deepseek_ocr/runtime/cuda_stream.h
families/deepseek_ocr/runtime/decoded_image.h
families/deepseek_ocr/runtime/image_preprocessor.cpp
families/deepseek_ocr/runtime/image_preprocessor.h
families/deepseek_ocr/runtime/image_transform_helper.h
families/deepseek_ocr/runtime/inference_state.h
families/deepseek_ocr/runtime/kv_cache.cpp
families/deepseek_ocr/runtime/kv_cache.h
families/deepseek_ocr/runtime/pipeline.cpp
families/deepseek_ocr/runtime/pipeline.h
families/deepseek_ocr/runtime/plugin.cpp
families/deepseek_ocr/runtime/plugin_helpers.cpp
families/deepseek_ocr/runtime/plugin_helpers.h
families/deepseek_ocr/runtime/runtime_config.cpp
families/deepseek_ocr/runtime/runtime_config.h
families/deepseek_ocr/runtime/sampler.cpp
families/deepseek_ocr/runtime/sampler.h
families/deepseek_ocr/runtime/tensor_names.h
families/deepseek_ocr/runtime/tokenizer.h
families/deepseek_ocr/support.py
families/deepseek_ocr/tests/__init__.py
families/deepseek_ocr/tests/cpp/test_runtime_config.cpp
families/deepseek_ocr/tests/data/deepseek_ocr_golden.json
families/deepseek_ocr/tests/data/deepseek_ocr_l0_golden.json
families/deepseek_ocr/tests/manifests/deepseek-ocr-l0-tp2.json
families/deepseek_ocr/tests/manifests/deepseek-ocr-l0.json
families/deepseek_ocr/tests/manifests/deepseek-ocr.json
families/deepseek_ocr/tests/test_e2e.py
families/deepseek_ocr/tests/test_vision_oracle.py
families/deepseek_ocr/tests/thresholds/deepseek-ocr-l0-tp2.json
families/deepseek_ocr/tests/thresholds/deepseek-ocr-l0.json
families/deepseek_ocr/tests/thresholds/deepseek-ocr.json
families/deepseek_ocr/tests/vision_oracle.py
families/deepseek_ocr/tp_builder.py
families/deepseek_ocr/utils.py
families/deepseek_v2/__init__.py
families/deepseek_v2/checkpoint_mapper.py
families/deepseek_v2/config.py
families/deepseek_v2/default_decoder.py
families/deepseek_v2/default_dual_profile_decoder.py
families/deepseek_v2/graph_blocks.py
families/deepseek_v2/graph_ops.py
families/deepseek_v2/model.py
families/deepseek_v2/moe_routing.py
families/deepseek_v2/parallel.py
families/deepseek_v2/runtime/CMakeLists.txt
families/deepseek_v2/runtime/bpe_tokenizer.cpp
families/deepseek_v2/runtime/chat_templates.cpp
families/deepseek_v2/runtime/chat_templates.h
families/deepseek_v2/runtime/distributed_runtime.cpp
families/deepseek_v2/runtime/distributed_runtime.h
families/deepseek_v2/runtime/inference_state.h
families/deepseek_v2/runtime/kv_cache.cpp
families/deepseek_v2/runtime/kv_cache.h
families/deepseek_v2/runtime/pipeline.cpp
families/deepseek_v2/runtime/pipeline.h
families/deepseek_v2/runtime/plugin.cpp
families/deepseek_v2/runtime/plugin_helpers.cpp
families/deepseek_v2/runtime/plugin_helpers.h
families/deepseek_v2/runtime/sampler.cpp
families/deepseek_v2/runtime/sampler.h
families/deepseek_v2/runtime/tensor_names.h
families/deepseek_v2/runtime/tokenizer.h
families/deepseek_v2/support.py
families/deepseek_v2/tests/__init__.py
families/deepseek_v2/tests/cpp/test_parallel_kv_contract.cpp
families/deepseek_v2/tests/manifests/deepseek-v2-lite-tp4.json
families/deepseek_v2/tests/manifests/deepseek-v2-lite.json
families/deepseek_v2/tests/manifests/deepseek-v2-tiny-tp2.json
families/deepseek_v2/tests/manifests/deepseek-v2-tiny.json
families/deepseek_v2/tests/test_e2e.py
families/deepseek_v2/tests/thresholds/deepseek-v2-lite-tp4.json
families/deepseek_v2/tests/thresholds/deepseek-v2-lite.json
families/deepseek_v2/tests/thresholds/deepseek-v2-tiny-tp2.json
families/deepseek_v2/tests/thresholds/deepseek-v2-tiny.json
families/deepseek_v2/tp_builder.py
families/deepseek_v2/utils.py
families/dinov3/__init__.py
families/dinov3/checkpoint_mapper.py
families/dinov3/config.py
families/dinov3/convnext_builder.py
families/dinov3/graph_ops.py
families/dinov3/model.py
families/dinov3/requirements.txt
families/dinov3/runtime/CMakeLists.txt
families/dinov3/runtime/image_preprocess.cpp
families/dinov3/runtime/image_preprocess.h
families/dinov3/runtime/pipeline.cpp
families/dinov3/runtime/pipeline.h
families/dinov3/runtime/plugin.cpp
families/dinov3/support.py
families/dinov3/tests/__init__.py
families/dinov3/tests/data/validation.json
families/dinov3/tests/manifests/dinov3-convnext-tiny-pretrain-lvd1689m.json
families/dinov3/tests/manifests/dinov3-vits16-pretrain-lvd1689m.json
families/dinov3/tests/manifests/dinov3-vits16-timm-l0.json
families/dinov3/tests/test_e2e.py
families/dinov3/tests/thresholds/dinov3-convnext-tiny-pretrain-lvd1689m.json
families/dinov3/tests/thresholds/dinov3-vits16-pretrain-lvd1689m.json
families/dinov3/tests/thresholds/dinov3-vits16-timm-l0.json
families/distilbert/__init__.py
families/distilbert/checkpoint_mapper.py
families/distilbert/config.py
families/distilbert/encoder_builder.py
families/distilbert/graph_ops.py
families/distilbert/model.py
families/distilbert/parallel.py
families/distilbert/runtime/CMakeLists.txt
families/distilbert/runtime/pipeline.cpp
families/distilbert/runtime/pipeline.h
families/distilbert/runtime/plugin.cpp
families/distilbert/runtime/plugin_helpers.cpp
families/distilbert/runtime/plugin_helpers.h
families/distilbert/runtime/tokenizer.h
families/distilbert/runtime/wordpiece_tokenizer.cpp
families/distilbert/support.py
families/distilbert/tests/__init__.py
families/distilbert/tests/manifests/distilbert-base-uncased-tp4.json
families/distilbert/tests/manifests/distilbert-base-uncased.json
families/distilbert/tests/test_e2e.py
families/distilbert/tests/thresholds/distilbert-base-uncased-tp4.json
families/distilbert/tests/thresholds/distilbert-base-uncased.json
families/distilbert/tp_builder.py
families/dpr/__init__.py
families/dpr/checkpoint_mapper.py
families/dpr/config.py
families/dpr/encoder_builder.py
families/dpr/graph_ops.py
families/dpr/model.py
families/dpr/parallel.py
families/dpr/runtime/CMakeLists.txt
families/dpr/runtime/pipeline.cpp
families/dpr/runtime/pipeline.h
families/dpr/runtime/plugin.cpp
families/dpr/runtime/plugin_helpers.cpp
families/dpr/runtime/plugin_helpers.h
families/dpr/runtime/tokenizer.h
families/dpr/runtime/wordpiece_tokenizer.cpp
families/dpr/support.py
families/dpr/tests/__init__.py
families/dpr/tests/manifests/dpr-ctx-encoder-tp4.json
families/dpr/tests/manifests/dpr-ctx-encoder.json
families/dpr/tests/test_e2e.py
families/dpr/tests/thresholds/dpr-ctx-encoder-tp4.json
families/dpr/tests/thresholds/dpr-ctx-encoder.json
families/dpr/tp_builder.py
families/eagle_vlm/__init__.py
families/eagle_vlm/checkpoint_mapper.py
families/eagle_vlm/config.py
families/eagle_vlm/graph_blocks.py
families/eagle_vlm/graph_ops.py
families/eagle_vlm/model.py
families/eagle_vlm/parallel.py
families/eagle_vlm/runtime/CMakeLists.txt
families/eagle_vlm/runtime/bpe_tokenizer.cpp
families/eagle_vlm/runtime/pipeline.cpp
families/eagle_vlm/runtime/pipeline.h
families/eagle_vlm/runtime/plugin.cpp
families/eagle_vlm/runtime/plugin_helpers.cpp
families/eagle_vlm/runtime/plugin_helpers.h
families/eagle_vlm/runtime/tokenizer.h
families/eagle_vlm/support.py
families/eagle_vlm/tests/__init__.py
families/eagle_vlm/tests/manifests/nemotron-embed-vl-1b-v2-tp4.json
families/eagle_vlm/tests/manifests/nemotron-embed-vl-1b-v2.json
families/eagle_vlm/tests/manifests/nemotron-rerank-vl-1b-v2-tp4.json
families/eagle_vlm/tests/manifests/nemotron-rerank-vl-1b-v2.json
families/eagle_vlm/tests/test_e2e.py
families/eagle_vlm/tests/thresholds/nemotron-embed-vl-1b-v2-tp4.json
families/eagle_vlm/tests/thresholds/nemotron-embed-vl-1b-v2.json
families/eagle_vlm/tests/thresholds/nemotron-rerank-vl-1b-v2-tp4.json
families/eagle_vlm/tests/thresholds/nemotron-rerank-vl-1b-v2.json
families/eagle_vlm/tp_builder.py
families/electra/__init__.py
families/electra/checkpoint_mapper.py
families/electra/config.py
families/electra/encoder_builder.py
families/electra/graph_ops.py
families/electra/model.py
families/electra/parallel.py
families/electra/runtime/CMakeLists.txt
families/electra/runtime/pipeline.cpp
families/electra/runtime/pipeline.h
families/electra/runtime/plugin.cpp
families/electra/runtime/plugin_helpers.cpp
families/electra/runtime/plugin_helpers.h
families/electra/runtime/tokenizer.h
families/electra/runtime/wordpiece_tokenizer.cpp
families/electra/support.py
families/electra/tests/__init__.py
families/electra/tests/manifests/electra-base-discriminator-tp4.json
families/electra/tests/manifests/electra-base-discriminator.json
families/electra/tests/test_e2e.py
families/electra/tests/thresholds/electra-base-discriminator-tp4.json
families/electra/tests/thresholds/electra-base-discriminator.json
families/electra/tp_builder.py
families/elf_flow/__init__.py
families/elf_flow/builder.py
families/elf_flow/checkpoint_mapper.py
families/elf_flow/config.py
families/elf_flow/graph_ops.py
families/elf_flow/model.py
families/elf_flow/model_config.py
families/elf_flow/requirements.txt
families/elf_flow/runtime/CMakeLists.txt
families/elf_flow/runtime/pipeline.cpp
families/elf_flow/runtime/pipeline.h
families/elf_flow/runtime/plugin.cpp
families/elf_flow/runtime/plugin_helpers.cpp
families/elf_flow/runtime/plugin_helpers.h
families/elf_flow/runtime/runtime_config.h
families/elf_flow/runtime/sampling_kernels.cu
families/elf_flow/runtime/sampling_kernels.h
families/elf_flow/runtime/tokenizer.h
families/elf_flow/runtime/unigram_tokenizer.cpp
families/elf_flow/support.py
families/elf_flow/t5_encoder_builder.py
families/elf_flow/tests/__init__.py
families/elf_flow/tests/cpp/test_elf_flow_runtime_config.cpp
families/elf_flow/tests/data/elf-b-de-en-replay/artifact.json
families/elf_flow/tests/data/elf-b-de-en-replay/condition_latents.f32
families/elf_flow/tests/data/elf-b-de-en-replay/condition_mask.f32
families/elf_flow/tests/data/elf-b-de-en-replay/expected.jsonl
families/elf_flow/tests/data/elf-b-de-en-replay/initial_latents.f32
families/elf_flow/tests/data/elf-b-de-en-replay/sampling_steps.f32
families/elf_flow/tests/data/elf-b-owt-replay/artifact.json
families/elf_flow/tests/data/elf-b-owt-replay/expected.jsonl
families/elf_flow/tests/data/elf-b-owt-replay/initial_latents.f32
families/elf_flow/tests/data/elf-b-owt-replay/sampling_steps.f32
families/elf_flow/tests/data/elf-b-xsum-replay/artifact.json
families/elf_flow/tests/data/elf-b-xsum-replay/expected.jsonl
families/elf_flow/tests/data/elf-b-xsum-replay/initial_latents.f32
families/elf_flow/tests/data/elf-b-xsum-replay/sampling_steps.f32
families/elf_flow/tests/manifests/elf-b-de-en-l0.json
families/elf_flow/tests/manifests/elf-b-owt-l0.json
families/elf_flow/tests/manifests/elf-b-xsum-l0.json
families/elf_flow/tests/test_e2e.py
families/elf_flow/tests/thresholds/elf-b-de-en-l0.json
families/elf_flow/tests/thresholds/elf-b-owt-l0.json
families/elf_flow/tests/thresholds/elf-b-xsum-l0.json
families/falcon/__init__.py
families/falcon/checkpoint_mapper.py
families/falcon/config.py
families/falcon/dual_profile_decoder_builder.py
families/falcon/dual_profile_decoder_tp_builder.py
families/falcon/graph_blocks.py
families/falcon/graph_ops.py
families/falcon/model.py
families/falcon/parallel.py
families/falcon/runtime/CMakeLists.txt
families/falcon/runtime/bpe_tokenizer.cpp
families/falcon/runtime/chat_templates.cpp
families/falcon/runtime/chat_templates.h
families/falcon/runtime/distributed_runtime.cpp
families/falcon/runtime/distributed_runtime.h
families/falcon/runtime/inference_state.h
families/falcon/runtime/kv_cache.cpp
families/falcon/runtime/kv_cache.h
families/falcon/runtime/pipeline.cpp
families/falcon/runtime/pipeline.h
families/falcon/runtime/plugin.cpp
families/falcon/runtime/plugin_helpers.cpp
families/falcon/runtime/plugin_helpers.h
families/falcon/runtime/sampler.cpp
families/falcon/runtime/sampler.h
families/falcon/runtime/tensor_names.h
families/falcon/runtime/tokenizer.h
families/falcon/standard_decoder_builder.py
families/falcon/support.py
families/falcon/tests/__init__.py
families/falcon/tests/manifests/falcon-rw-1b-tp4.json
families/falcon/tests/manifests/falcon-rw-1b.json
families/falcon/tests/test_e2e.py
families/falcon/tests/thresholds/falcon-rw-1b-tp4.json
families/falcon/tests/thresholds/falcon-rw-1b.json
families/fast_foundation_stereo/__init__.py
families/fast_foundation_stereo/builder.py
families/fast_foundation_stereo/model.py
families/fast_foundation_stereo/native_feature.py
families/fast_foundation_stereo/native_graph.py
families/fast_foundation_stereo/native_plugin_builder.py
families/fast_foundation_stereo/native_post.py
families/fast_foundation_stereo/prepare_model.py
families/fast_foundation_stereo/requirements.txt
families/fast_foundation_stereo/runtime/CMakeLists.txt
families/fast_foundation_stereo/runtime/native_plugins/CMakeLists.txt
families/fast_foundation_stereo/runtime/native_plugins/full_volume_leaky_plugin.cu
families/fast_foundation_stereo/runtime/native_plugins/geometry_volume_convc1_plugin.cu
families/fast_foundation_stereo/runtime/native_plugins/gwc_plugin.cu
families/fast_foundation_stereo/runtime/native_plugins/plugin_creators.cpp
families/fast_foundation_stereo/runtime/native_plugins/plugins.h
families/fast_foundation_stereo/runtime/native_plugins/post8_sum_plugin.cu
families/fast_foundation_stereo/runtime/native_plugins/spatial_attention_reduce_plugin.cu
families/fast_foundation_stereo/runtime/plugin.cpp
families/fast_foundation_stereo/runtime/stereo_pipeline.cpp
families/fast_foundation_stereo/runtime/stereo_pipeline.h
families/fast_foundation_stereo/support.py
families/fast_foundation_stereo/tests/__init__.py
families/fast_foundation_stereo/tests/accuracy.py
families/fast_foundation_stereo/tests/cpp/test_fast_foundation_stereo_preprocess.cpp
families/fast_foundation_stereo/tests/data/README.md
families/fast_foundation_stereo/tests/input_contract.py
families/fast_foundation_stereo/tests/manifests/fast-foundation-stereo.json
families/fast_foundation_stereo/tests/prepare_middlebury_q.py
families/fast_foundation_stereo/tests/reference-source.json
families/fast_foundation_stereo/tests/test_e2e.py
families/fast_foundation_stereo/tests/test_prepare_middlebury_q.py
families/fast_foundation_stereo/tests/test_task_accuracy.py
families/fast_foundation_stereo/tests/thresholds/fast-foundation-stereo-middlebury-q.json
families/fast_foundation_stereo/tests/thresholds/fast-foundation-stereo.json
families/flux/__init__.py
families/flux/checkpoint_mapper.py
families/flux/clip_encoder_builder.py
families/flux/config.py
families/flux/data/flux2-fp8-scales.json
families/flux/flux2_dit_builder.py
families/flux/flux2_dit_tp_builder.py
families/flux/flux_dit_builder.py
families/flux/flux_dit_cp_builder.py
families/flux/flux_dit_tp_builder.py
families/flux/flux_vae_builder.py
families/flux/graph_ops.py
families/flux/mistral_encoder_builder.py
families/flux/model.py
families/flux/parallel.py
families/flux/requirements.txt
families/flux/runtime/CMakeLists.txt
families/flux/runtime/bpe_tokenizer.cpp
families/flux/runtime/diffusion_helpers.cpp
families/flux/runtime/diffusion_helpers.h
families/flux/runtime/distributed_runtime.cpp
families/flux/runtime/distributed_runtime.h
families/flux/runtime/flux_batch_utils.h
families/flux/runtime/flux_clip_helpers.h
families/flux/runtime/flux_denoising_step_seam.h
families/flux/runtime/flux_diffusion_types.h
families/flux/runtime/flux_generation_plan.h
families/flux/runtime/flux_rope_helpers.h
families/flux/runtime/flux_scheduler_helpers.h
families/flux/runtime/flux_text_helpers.h
families/flux/runtime/gpu_matmul.cpp
families/flux/runtime/gpu_matmul.h
families/flux/runtime/pipeline.cpp
families/flux/runtime/pipeline.h
families/flux/runtime/plugin.cpp
families/flux/runtime/plugin_helpers.cpp
families/flux/runtime/plugin_helpers.h
families/flux/runtime/preprocessor_weights_helpers.h
families/flux/runtime/tokenizer.h
families/flux/runtime/unigram_tokenizer.cpp
families/flux/support.py
families/flux/t5_encoder_builder.py
families/flux/tests/__init__.py
families/flux/tests/data/flux2-fp8-scales.json
families/flux/tests/manifests/flux-2-dev-fp8-l0-tp4.json
families/flux/tests/manifests/flux-2-dev-fp8-l0.json
families/flux/tests/manifests/flux-2-dev-fp8.json
families/flux/tests/manifests/flux-2-dev-l0-tp4.json
families/flux/tests/manifests/flux-2-dev-l0.json
families/flux/tests/manifests/flux-2-dev.json
families/flux/tests/manifests/flux-schnell-l0-batch2.json
families/flux/tests/manifests/flux-schnell-l0-cp4.json
families/flux/tests/manifests/flux-schnell-l0-tp4.json
families/flux/tests/manifests/flux-schnell-l0.json
families/flux/tests/manifests/flux-schnell.json
families/flux/tests/test_context_parallel.py
families/flux/tests/test_e2e.py
families/flux/tests/thresholds/flux-2-dev-fp8-l0-tp4.json
families/flux/tests/thresholds/flux-2-dev-fp8-l0.json
families/flux/tests/thresholds/flux-2-dev-fp8.json
families/flux/tests/thresholds/flux-2-dev-l0-tp4.json
families/flux/tests/thresholds/flux-2-dev-l0.json
families/flux/tests/thresholds/flux-2-dev.json
families/flux/tests/thresholds/flux-schnell-l0-batch2.json
families/flux/tests/thresholds/flux-schnell-l0-cp4.json
families/flux/tests/thresholds/flux-schnell-l0-tp4.json
families/flux/tests/thresholds/flux-schnell-l0.json
families/flux/tests/thresholds/flux-schnell.json
families/flux/timing.py
families/fnet/__init__.py
families/fnet/checkpoint_mapper.py
families/fnet/config.py
families/fnet/fnet_encoder_builder.py
families/fnet/graph_ops.py
families/fnet/model.py
families/fnet/parallel.py
families/fnet/runtime/CMakeLists.txt
families/fnet/runtime/bpe_tokenizer.cpp
families/fnet/runtime/pipeline.cpp
families/fnet/runtime/pipeline.h
families/fnet/runtime/plugin.cpp
families/fnet/runtime/plugin_helpers.cpp
families/fnet/runtime/plugin_helpers.h
families/fnet/runtime/tokenizer.h
families/fnet/support.py
families/fnet/tests/__init__.py
families/fnet/tests/manifests/fnet-base-tp4.json
families/fnet/tests/manifests/fnet-base.json
families/fnet/tests/test_e2e.py
families/fnet/tests/thresholds/fnet-base-tp4.json
families/fnet/tests/thresholds/fnet-base.json
families/fnet/tp_builder.py
families/foundationpose/__init__.py
families/foundationpose/builder.py
families/foundationpose/model.py
families/foundationpose/requirements.txt
families/foundationpose/runtime/CMakeLists.txt
families/foundationpose/runtime/pipeline.cpp
families/foundationpose/runtime/pipeline.h
families/foundationpose/runtime/plugin.cpp
families/foundationpose/support.py
families/foundationpose/tests/__init__.py
families/foundationpose/tests/cpp/qualification.cpp
families/foundationpose/tests/cpp/test_pipeline.cpp
families/foundationpose/tests/manifests/foundationpose-ngc-1.0.1.json
families/foundationpose/tests/official_reference.py
families/foundationpose/tests/test_e2e.py
families/foundationpose/tests/test_model.py
families/foundationpose/tests/thresholds/foundationpose-ngc-1.0.1-synthetic-crops.json
families/gemma/__init__.py
families/gemma/checkpoint_mapper.py
families/gemma/config.py
families/gemma/dual_profile_decoder_builder.py
families/gemma/dual_profile_decoder_tp_builder.py
families/gemma/graph_blocks.py
families/gemma/graph_ops.py
families/gemma/model.py
families/gemma/parallel.py
families/gemma/runtime/CMakeLists.txt
families/gemma/runtime/bpe_tokenizer.cpp
families/gemma/runtime/chat_templates.cpp
families/gemma/runtime/chat_templates.h
families/gemma/runtime/distributed_runtime.cpp
families/gemma/runtime/distributed_runtime.h
families/gemma/runtime/inference_state.h
families/gemma/runtime/kv_cache.cpp
families/gemma/runtime/kv_cache.h
families/gemma/runtime/pipeline.cpp
families/gemma/runtime/pipeline.h
families/gemma/runtime/plugin.cpp
families/gemma/runtime/plugin_helpers.cpp
families/gemma/runtime/plugin_helpers.h
families/gemma/runtime/sampler.cpp
families/gemma/runtime/sampler.h
families/gemma/runtime/tensor_names.h
families/gemma/runtime/tokenizer.h
families/gemma/standard_decoder_builder.py
families/gemma/support.py
families/gemma/tests/__init__.py
families/gemma/tests/cpp/test_gemma_pipeline.cpp
families/gemma/tests/cpp/test_gemma_sampler.cpp
families/gemma/tests/manifests/gemma-2-2b-tp4.json
families/gemma/tests/manifests/gemma-2-2b.json
families/gemma/tests/test_e2e.py
families/gemma/tests/thresholds/gemma-2-2b-tp4.json
families/gemma/tests/thresholds/gemma-2-2b.json
families/glm/__init__.py
families/glm/checkpoint_mapper.py
families/glm/config.py
families/glm/default_decoder.py
families/glm/default_dual_profile_decoder.py
families/glm/dual_profile_decoder_tp_builder.py
families/glm/graph_blocks.py
families/glm/graph_ops.py
families/glm/model.py
families/glm/parallel.py
families/glm/runtime/CMakeLists.txt
families/glm/runtime/bpe_tokenizer.cpp
families/glm/runtime/chat_templates.cpp
families/glm/runtime/chat_templates.h
families/glm/runtime/distributed_runtime.cpp
families/glm/runtime/distributed_runtime.h
families/glm/runtime/inference_state.h
families/glm/runtime/kv_cache.cpp
families/glm/runtime/kv_cache.h
families/glm/runtime/pipeline.cpp
families/glm/runtime/pipeline.h
families/glm/runtime/plugin.cpp
families/glm/runtime/plugin_helpers.cpp
families/glm/runtime/plugin_helpers.h
families/glm/runtime/sampler.cpp
families/glm/runtime/sampler.h
families/glm/runtime/tensor_names.h
families/glm/runtime/tokenizer.h
families/glm/support.py
families/glm/tests/__init__.py
families/glm/tests/manifests/glm-4-9b-l0-tp2.json
families/glm/tests/manifests/glm-4-9b-l0.json
families/glm/tests/manifests/glm-4-9b.json
families/glm/tests/test_e2e.py
families/glm/tests/thresholds/glm-4-9b-l0-tp2.json
families/glm/tests/thresholds/glm-4-9b-l0.json
families/glm/tests/thresholds/glm-4-9b.json
families/glm/utils.py
families/gpt2/__init__.py
families/gpt2/checkpoint_mapper.py
families/gpt2/config.py
families/gpt2/default_decoder.py
families/gpt2/default_dual_profile_decoder.py
families/gpt2/default_dual_profile_decoder_tp.py
families/gpt2/graph_blocks.py
families/gpt2/graph_ops.py
families/gpt2/model.py
families/gpt2/parallel.py
families/gpt2/runtime/CMakeLists.txt
families/gpt2/runtime/bpe_tokenizer.cpp
families/gpt2/runtime/chat_templates.cpp
families/gpt2/runtime/chat_templates.h
families/gpt2/runtime/distributed_runtime.cpp
families/gpt2/runtime/distributed_runtime.h
families/gpt2/runtime/inference_state.h
families/gpt2/runtime/kv_cache.cpp
families/gpt2/runtime/kv_cache.h
families/gpt2/runtime/pipeline.cpp
families/gpt2/runtime/pipeline.h
families/gpt2/runtime/plugin.cpp
families/gpt2/runtime/plugin_helpers.cpp
families/gpt2/runtime/plugin_helpers.h
families/gpt2/runtime/sampler.cpp
families/gpt2/runtime/sampler.h
families/gpt2/runtime/tensor_names.h
families/gpt2/runtime/tokenizer.h
families/gpt2/support.py
families/gpt2/tests/__init__.py
families/gpt2/tests/cpp/test_parallel_kv_contract.cpp
families/gpt2/tests/manifests/distilgpt2.json
families/gpt2/tests/manifests/gpt2-125m-tp4.json
families/gpt2/tests/manifests/gpt2-125m.json
families/gpt2/tests/test_e2e.py
families/gpt2/tests/thresholds/distilgpt2.json
families/gpt2/tests/thresholds/gpt2-125m-tp4.json
families/gpt2/tests/thresholds/gpt2-125m.json
families/gpt2/utils.py
families/gpt_neo/__init__.py
families/gpt_neo/attention_contract.py
families/gpt_neo/checkpoint_mapper.py
families/gpt_neo/config.py
families/gpt_neo/default_decoder.py
families/gpt_neo/default_dual_profile_decoder.py
families/gpt_neo/default_dual_profile_decoder_tp.py
families/gpt_neo/graph_blocks.py
families/gpt_neo/graph_ops.py
families/gpt_neo/model.py
families/gpt_neo/parallel.py
families/gpt_neo/runtime/CMakeLists.txt
families/gpt_neo/runtime/bpe_tokenizer.cpp
families/gpt_neo/runtime/chat_templates.cpp
families/gpt_neo/runtime/chat_templates.h
families/gpt_neo/runtime/distributed_runtime.cpp
families/gpt_neo/runtime/distributed_runtime.h
families/gpt_neo/runtime/inference_state.h
families/gpt_neo/runtime/kv_cache.cpp
families/gpt_neo/runtime/kv_cache.h
families/gpt_neo/runtime/pipeline.cpp
families/gpt_neo/runtime/pipeline.h
families/gpt_neo/runtime/plugin.cpp
families/gpt_neo/runtime/plugin_helpers.cpp
families/gpt_neo/runtime/plugin_helpers.h
families/gpt_neo/runtime/sampler.cpp
families/gpt_neo/runtime/sampler.h
families/gpt_neo/runtime/tensor_names.h
families/gpt_neo/runtime/tokenizer.h
families/gpt_neo/support.py
families/gpt_neo/tests/__init__.py
families/gpt_neo/tests/manifests/gpt-neo-125m-tp4.json
families/gpt_neo/tests/manifests/gpt-neo-125m.json
families/gpt_neo/tests/test_e2e.py
families/gpt_neo/tests/thresholds/gpt-neo-125m-tp4.json
families/gpt_neo/tests/thresholds/gpt-neo-125m.json
families/gpt_neo/utils.py
families/gpt_neox/__init__.py
families/gpt_neox/checkpoint_mapper.py
families/gpt_neox/config.py
families/gpt_neox/default_decoder.py
families/gpt_neox/default_dual_profile_decoder.py
families/gpt_neox/default_dual_profile_decoder_tp.py
families/gpt_neox/graph_blocks.py
families/gpt_neox/graph_ops.py
families/gpt_neox/model.py
families/gpt_neox/parallel.py
families/gpt_neox/runtime/CMakeLists.txt
families/gpt_neox/runtime/bpe_tokenizer.cpp
families/gpt_neox/runtime/chat_templates.cpp
families/gpt_neox/runtime/chat_templates.h
families/gpt_neox/runtime/distributed_runtime.cpp
families/gpt_neox/runtime/distributed_runtime.h
families/gpt_neox/runtime/inference_state.h
families/gpt_neox/runtime/kv_cache.cpp
families/gpt_neox/runtime/kv_cache.h
families/gpt_neox/runtime/pipeline.cpp
families/gpt_neox/runtime/pipeline.h
families/gpt_neox/runtime/plugin.cpp
families/gpt_neox/runtime/plugin_helpers.cpp
families/gpt_neox/runtime/plugin_helpers.h
families/gpt_neox/runtime/sampler.cpp
families/gpt_neox/runtime/sampler.h
families/gpt_neox/runtime/tensor_names.h
families/gpt_neox/runtime/tokenizer.h
families/gpt_neox/support.py
families/gpt_neox/tests/__init__.py
families/gpt_neox/tests/cpp/test_gpt_neox_attention_mask.cpp
families/gpt_neox/tests/manifests/pythia-70m-tp4.json
families/gpt_neox/tests/manifests/pythia-70m.json
families/gpt_neox/tests/test_e2e.py
families/gpt_neox/tests/thresholds/pythia-70m-tp4.json
families/gpt_neox/tests/thresholds/pythia-70m.json
families/gpt_neox/utils.py
families/gpt_oss/__init__.py
families/gpt_oss/checkpoint_mapper.py
families/gpt_oss/config.py
families/gpt_oss/default_decoder.py
families/gpt_oss/default_dual_profile_decoder.py
families/gpt_oss/graph_blocks.py
families/gpt_oss/graph_ops.py
families/gpt_oss/model.py
families/gpt_oss/parallel.py
families/gpt_oss/requirements.txt
families/gpt_oss/runtime/CMakeLists.txt
families/gpt_oss/runtime/bpe_tokenizer.cpp
families/gpt_oss/runtime/chat_templates.cpp
families/gpt_oss/runtime/chat_templates.h
families/gpt_oss/runtime/distributed_runtime.cpp
families/gpt_oss/runtime/distributed_runtime.h
families/gpt_oss/runtime/inference_state.h
families/gpt_oss/runtime/kv_cache.cpp
families/gpt_oss/runtime/kv_cache.h
families/gpt_oss/runtime/pipeline.cpp
families/gpt_oss/runtime/pipeline.h
families/gpt_oss/runtime/plugin.cpp
families/gpt_oss/runtime/plugin_helpers.cpp
families/gpt_oss/runtime/plugin_helpers.h
families/gpt_oss/runtime/sampler.cpp
families/gpt_oss/runtime/sampler.h
families/gpt_oss/runtime/tensor_names.h
families/gpt_oss/runtime/tokenizer.h
families/gpt_oss/support.py
families/gpt_oss/tests/__init__.py
families/gpt_oss/tests/manifests/gpt-oss-20b-l0-tp4.json
families/gpt_oss/tests/manifests/gpt-oss-20b-l0.json
families/gpt_oss/tests/manifests/gpt-oss-20b.json
families/gpt_oss/tests/test_e2e.py
families/gpt_oss/tests/thresholds/gpt-oss-20b-l0-tp4.json
families/gpt_oss/tests/thresholds/gpt-oss-20b-l0.json
families/gpt_oss/tests/thresholds/gpt-oss-20b.json
families/gpt_oss/tp_builder.py
families/gpt_oss/utils.py
families/granite/__init__.py
families/granite/checkpoint_mapper.py
families/granite/config.py
families/granite/default_decoder.py
families/granite/default_dual_profile_decoder.py
families/granite/dual_profile_decoder_tp_builder.py
families/granite/graph_blocks.py
families/granite/graph_ops.py
families/granite/model.py
families/granite/parallel.py
families/granite/runtime/CMakeLists.txt
families/granite/runtime/bpe_tokenizer.cpp
families/granite/runtime/chat_templates.cpp
families/granite/runtime/chat_templates.h
families/granite/runtime/distributed_runtime.cpp
families/granite/runtime/distributed_runtime.h
families/granite/runtime/inference_state.h
families/granite/runtime/kv_cache.cpp
families/granite/runtime/kv_cache.h
families/granite/runtime/pipeline.cpp
families/granite/runtime/pipeline.h
families/granite/runtime/plugin.cpp
families/granite/runtime/plugin_helpers.cpp
families/granite/runtime/plugin_helpers.h
families/granite/runtime/sampler.cpp
families/granite/runtime/sampler.h
families/granite/runtime/tensor_names.h
families/granite/runtime/tokenizer.h
families/granite/support.py
families/granite/tests/__init__.py
families/granite/tests/manifests/granite-3.1-2b-tp4.json
families/granite/tests/manifests/granite-3.1-2b.json
families/granite/tests/test_e2e.py
families/granite/tests/thresholds/granite-3.1-2b-tp4.json
families/granite/tests/thresholds/granite-3.1-2b.json
families/granite/utils.py
families/internlm/__init__.py
families/internlm/checkpoint_mapper.py
families/internlm/config.py
families/internlm/default_decoder.py
families/internlm/default_dual_profile_decoder.py
families/internlm/dual_profile_decoder_tp_builder.py
families/internlm/graph_blocks.py
families/internlm/graph_ops.py
families/internlm/model.py
families/internlm/parallel.py
families/internlm/requirements.txt
families/internlm/runtime/CMakeLists.txt
families/internlm/runtime/bpe_tokenizer.cpp
families/internlm/runtime/chat_templates.cpp
families/internlm/runtime/chat_templates.h
families/internlm/runtime/distributed_runtime.cpp
families/internlm/runtime/distributed_runtime.h
families/internlm/runtime/inference_state.h
families/internlm/runtime/kv_cache.cpp
families/internlm/runtime/kv_cache.h
families/internlm/runtime/pipeline.cpp
families/internlm/runtime/pipeline.h
families/internlm/runtime/plugin.cpp
families/internlm/runtime/plugin_helpers.cpp
families/internlm/runtime/plugin_helpers.h
families/internlm/runtime/sampler.cpp
families/internlm/runtime/sampler.h
families/internlm/runtime/tensor_names.h
families/internlm/runtime/tokenizer.h
families/internlm/support.py
families/internlm/tests/__init__.py
families/internlm/tests/cpp/test_internlm_chat_template.cpp
families/internlm/tests/cpp/test_internlm_pipeline.cpp
families/internlm/tests/manifests/internlm2-1.8b-tp4.json
families/internlm/tests/manifests/internlm2-1.8b.json
families/internlm/tests/test_e2e.py
families/internlm/tests/thresholds/internlm2-1.8b-tp4.json
families/internlm/tests/thresholds/internlm2-1.8b.json
families/internlm/tokenizer_json.py
families/internlm/utils.py
families/internvl/__init__.py
families/internvl/checkpoint_mapper.py
families/internvl/config.py
families/internvl/default_decoder.py
families/internvl/default_dual_profile_decoder.py
families/internvl/graph_blocks.py
families/internvl/graph_ops.py
families/internvl/internvit_vision_builder.py
families/internvl/model.py
families/internvl/parallel.py
families/internvl/requirements.txt
families/internvl/runtime/CMakeLists.txt
families/internvl/runtime/bpe_tokenizer.cpp
families/internvl/runtime/cuda_stream.h
families/internvl/runtime/decoded_image.h
families/internvl/runtime/distributed_runtime.cpp
families/internvl/runtime/distributed_runtime.h
families/internvl/runtime/image_preprocessor.cpp
families/internvl/runtime/image_preprocessor.h
families/internvl/runtime/image_transform_helper.h
families/internvl/runtime/inference_state.h
families/internvl/runtime/kv_cache.cpp
families/internvl/runtime/kv_cache.h
families/internvl/runtime/pipeline.cpp
families/internvl/runtime/pipeline.h
families/internvl/runtime/plugin.cpp
families/internvl/runtime/plugin_helpers.cpp
families/internvl/runtime/plugin_helpers.h
families/internvl/runtime/sampler.cpp
families/internvl/runtime/sampler.h
families/internvl/runtime/tensor_names.h
families/internvl/runtime/tokenizer.h
families/internvl/support.py
families/internvl/tests/__init__.py
families/internvl/tests/data/internvl3_8b_golden.json
families/internvl/tests/manifests/internvl3-2b-tp2.json
families/internvl/tests/manifests/internvl3-2b.json
families/internvl/tests/manifests/internvl3-8b-tp4.json
families/internvl/tests/manifests/internvl3-8b.json
families/internvl/tests/test_e2e.py
families/internvl/tests/test_tp_contract.py
families/internvl/tests/test_vision_oracle.py
families/internvl/tests/thresholds/internvl3-2b-tp2.json
families/internvl/tests/thresholds/internvl3-2b.json
families/internvl/tests/thresholds/internvl3-8b-tp4.json
families/internvl/tests/thresholds/internvl3-8b.json
families/internvl/tests/vision_oracle.py
families/internvl/tp_builder.py
families/internvl/utils.py
families/k2_horizon/__init__.py
families/k2_horizon/checkpoint_mapper.py
families/k2_horizon/config.py
families/k2_horizon/model.py
families/k2_horizon/native_kv_attention_builder.py
families/k2_horizon/requirements.txt
families/k2_horizon/runtime/CMakeLists.txt
families/k2_horizon/runtime/bpe_tokenizer.cpp
families/k2_horizon/runtime/inference_state.h
families/k2_horizon/runtime/kv_cache.cpp
families/k2_horizon/runtime/kv_cache.h
families/k2_horizon/runtime/pipeline.cpp
families/k2_horizon/runtime/pipeline.h
families/k2_horizon/runtime/plugin.cpp
families/k2_horizon/runtime/plugin_helpers.cpp
families/k2_horizon/runtime/plugin_helpers.h
families/k2_horizon/runtime/sampler.cpp
families/k2_horizon/runtime/sampler.h
families/k2_horizon/runtime/tensor_names.h
families/k2_horizon/runtime/tokenizer.h
families/k2_horizon/support.py
families/k2_horizon/tests/__init__.py
families/k2_horizon/tests/cpp/test_k2_horizon_sampler.cpp
families/k2_horizon/tests/debug_runner.py
families/k2_horizon/tests/manifests/k2-horizon-7b.json
families/k2_horizon/tests/test_e2e.py
families/k2_horizon/tests/test_model.py
families/k2_horizon/tests/thresholds/k2-horizon-7b.json
families/lance/__init__.py
families/lance/checkpoint_mapper.py
families/lance/config.py
families/lance/default_decoder.py
families/lance/default_dual_profile_decoder.py
families/lance/graph_blocks.py
families/lance/graph_ops.py
families/lance/model.py
families/lance/qwen_vl_vision_builder.py
families/lance/requirements.txt
families/lance/runtime/CMakeLists.txt
families/lance/runtime/bpe_tokenizer.cpp
families/lance/runtime/cuda_stream.h
families/lance/runtime/decoded_image.h
families/lance/runtime/image_preprocessor.cpp
families/lance/runtime/image_preprocessor.h
families/lance/runtime/image_transform_helper.h
families/lance/runtime/inference_state.h
families/lance/runtime/kv_cache.cpp
families/lance/runtime/kv_cache.h
families/lance/runtime/pipeline.cpp
families/lance/runtime/pipeline.h
families/lance/runtime/plugin.cpp
families/lance/runtime/plugin_helpers.cpp
families/lance/runtime/plugin_helpers.h
families/lance/runtime/sampler.cpp
families/lance/runtime/sampler.h
families/lance/runtime/tensor_names.h
families/lance/runtime/tokenizer.h
families/lance/support.py
families/lance/tests/__init__.py
families/lance/tests/manifests/lance-3b-x2t-image.json
families/lance/tests/official_reference.py
families/lance/tests/reference-source.json
families/lance/tests/test_e2e.py
families/lance/tests/test_official_reference.py
families/lance/tests/test_reference_dependencies.py
families/lance/tests/test_tokenizer_contract.py
families/lance/tests/test_vision_oracle.py
families/lance/tests/thresholds/lance-3b-x2t-image.json
families/lance/tests/vision_oracle.py
families/lance/utils.py
families/lerobot_act/__init__.py
families/lerobot_act/builder.py
families/lerobot_act/checkpoint.py
families/lerobot_act/model.py
families/lerobot_act/requirements.txt
families/lerobot_act/runtime/CMakeLists.txt
families/lerobot_act/runtime/pipeline.cpp
families/lerobot_act/runtime/pipeline.h
families/lerobot_act/runtime/plugin.cpp
families/lerobot_act/support.py
families/lerobot_act/tests/__init__.py
families/lerobot_act/tests/cpp/qualification.cpp
families/lerobot_act/tests/cpp/test_pipeline.cpp
families/lerobot_act/tests/data/recorded_observation/observation.state.f32
families/lerobot_act/tests/data/recorded_observation/recorded_observation.json
families/lerobot_act/tests/manifests/act-aloha-sim-transfer-cube.json
families/lerobot_act/tests/official_reference.py
families/lerobot_act/tests/performance_reference.py
families/lerobot_act/tests/prepare_recorded_observation.py
families/lerobot_act/tests/reference-source.json
families/lerobot_act/tests/test_e2e.py
families/lerobot_act/tests/test_model.py
families/lerobot_act/tests/test_recorded_observation.py
families/lerobot_act/tests/thresholds/act-aloha-sim-transfer-cube-recorded-episode-0-frame-0.json
families/lfm2/__init__.py
families/lfm2/checkpoint_mapper.py
families/lfm2/config.py
families/lfm2/model.py
families/lfm2/native_kv_attention_builder.py
families/lfm2/runtime/CMakeLists.txt
families/lfm2/runtime/bpe_tokenizer.cpp
families/lfm2/runtime/byte_level_decoder.cpp
families/lfm2/runtime/byte_level_decoder.h
families/lfm2/runtime/chat_templates.cpp
families/lfm2/runtime/chat_templates.h
families/lfm2/runtime/conv_state.cpp
families/lfm2/runtime/conv_state.h
families/lfm2/runtime/hybrid_state.cpp
families/lfm2/runtime/hybrid_state.h
families/lfm2/runtime/inference_state.h
families/lfm2/runtime/kv_cache.cpp
families/lfm2/runtime/kv_cache.h
families/lfm2/runtime/pipeline.cpp
families/lfm2/runtime/pipeline.h
families/lfm2/runtime/plugin.cpp
families/lfm2/runtime/plugin_helpers.cpp
families/lfm2/runtime/plugin_helpers.h
families/lfm2/runtime/pretokenizer.cpp
families/lfm2/runtime/pretokenizer.h
families/lfm2/runtime/sampler.cpp
families/lfm2/runtime/sampler.h
families/lfm2/runtime/tokenizer.h
families/lfm2/support.py
families/lfm2/tests/__init__.py
families/lfm2/tests/cpp/test_lfm2_byte_level_decoder.cpp
families/lfm2/tests/cpp/test_lfm2_chat_template.cpp
families/lfm2/tests/cpp/test_lfm2_pretokenizer.cpp
families/lfm2/tests/cpp/test_lfm2_runtime.cpp
families/lfm2/tests/cpp/test_lfm2_sampler.cpp
families/lfm2/tests/manifests/lfm2-1.2b.json
families/lfm2/tests/manifests/lfm2-2.6b.json
families/lfm2/tests/manifests/lfm2-350m-bf16-model-card.json
families/lfm2/tests/manifests/lfm2-350m-fp16.json
families/lfm2/tests/manifests/lfm2-700m.json
families/lfm2/tests/test_e2e.py
families/lfm2/tests/thresholds/lfm2-1.2b-greedy.json
families/lfm2/tests/thresholds/lfm2-2.6b-greedy.json
families/lfm2/tests/thresholds/lfm2-350m-bf16-model-card-sampling.json
families/lfm2/tests/thresholds/lfm2-350m-chat.json
families/lfm2/tests/thresholds/lfm2-350m-fp16.json
families/lfm2/tests/thresholds/lfm2-700m-greedy.json
families/lfm2/tests/validation/lfm2-350m-bf16-model-card.json
families/llama/__init__.py
families/llama/build_routing.py
families/llama/checkpoint_mapper.py
families/llama/config.py
families/llama/dual_profile_decoder_builder.py
families/llama/graph_blocks.py
families/llama/graph_ops.py
families/llama/model.py
families/llama/native_kv_attention_builder.py
families/llama/native_kv_contract.py
families/llama/runtime/CMakeLists.txt
families/llama/runtime/bpe_tokenizer.cpp
families/llama/runtime/chat_templates.cpp
families/llama/runtime/chat_templates.h
families/llama/runtime/inference_state.h
families/llama/runtime/kv_cache.cpp
families/llama/runtime/kv_cache.h
families/llama/runtime/pipeline.cpp
families/llama/runtime/pipeline.h
families/llama/runtime/plugin.cpp
families/llama/runtime/plugin_helpers.cpp
families/llama/runtime/plugin_helpers.h
families/llama/runtime/sampler.cpp
families/llama/runtime/sampler.h
families/llama/runtime/tensor_names.h
families/llama/runtime/tokenizer.h
families/llama/standard_decoder_builder.py
families/llama/support.py
families/llama/tests/__init__.py
families/llama/tests/cpp/native_kv_cache_contract_test.h
families/llama/tests/cpp/test_llama_chat_template.cpp
families/llama/tests/cpp/test_llama_native_kv_cache.cpp
families/llama/tests/manifests/falcon3-1b.json
families/llama/tests/manifests/minitron-4b-depth.json
families/llama/tests/manifests/minitron-4b-width-l0.json
families/llama/tests/manifests/minitron-4b-width-regression-native-kv-chunked-prefill.json
families/llama/tests/manifests/minitron-4b-width.json
families/llama/tests/manifests/nemotron-nano-4b.json
families/llama/tests/manifests/tinyllama-1.1b.json
families/llama/tests/runtime_receipt.py
families/llama/tests/test_e2e.py
families/llama/tests/test_runtime_receipt.py
families/llama/tests/thresholds/falcon3-1b.json
families/llama/tests/thresholds/minitron-4b-depth.json
families/llama/tests/thresholds/minitron-4b-width-l0.json
families/llama/tests/thresholds/minitron-4b-width-regression-native-kv-chunked-prefill.json
families/llama/tests/thresholds/minitron-4b-width.json
families/llama/tests/thresholds/nemotron-nano-4b.json
families/llama/tests/thresholds/tinyllama-1.1b.json
families/llama/utils.py
families/locateanything/__init__.py
families/locateanything/checkpoint_mapper.py
families/locateanything/config.py
families/locateanything/decoder_tp_builder.py
families/locateanything/default_decoder.py
families/locateanything/default_dual_profile_decoder.py
families/locateanything/graph_blocks.py
families/locateanything/graph_ops.py
families/locateanything/model.py
families/locateanything/onnx_vision_builder.py
families/locateanything/parallel.py
families/locateanything/requirements.txt
families/locateanything/runtime/CMakeLists.txt
families/locateanything/runtime/bpe_tokenizer.cpp
families/locateanything/runtime/cuda_stream.h
families/locateanything/runtime/decoded_image.h
families/locateanything/runtime/image_preprocessor.cpp
families/locateanything/runtime/image_preprocessor.h
families/locateanything/runtime/image_transform_helper.h
families/locateanything/runtime/inference_state.h
families/locateanything/runtime/kv_cache.cpp
families/locateanything/runtime/kv_cache.h
families/locateanything/runtime/pipeline.cpp
families/locateanything/runtime/pipeline.h
families/locateanything/runtime/plugin.cpp
families/locateanything/runtime/plugin_helpers.cpp
families/locateanything/runtime/plugin_helpers.h
families/locateanything/runtime/sampler.cpp
families/locateanything/runtime/sampler.h
families/locateanything/runtime/tensor_names.h
families/locateanything/runtime/tokenizer.h
families/locateanything/support.py
families/locateanything/tests/__init__.py
families/locateanything/tests/manifests/locateanything-3b.json
families/locateanything/tests/test_e2e.py
families/locateanything/tests/test_vision_oracle.py
families/locateanything/tests/thresholds/locateanything-3b-point.json
families/locateanything/tests/thresholds/locateanything-3b.json
families/locateanything/tests/vision_oracle.py
families/locateanything/utils.py
families/locateanything/vision_builder.py
families/ltx_video/__init__.py
families/ltx_video/checkpoint_mapper.py
families/ltx_video/config.py
families/ltx_video/graph_ops.py
families/ltx_video/ltx_dit_builder.py
families/ltx_video/ltx_vae_builder.py
families/ltx_video/model.py
families/ltx_video/requirements.txt
families/ltx_video/runtime/CMakeLists.txt
families/ltx_video/runtime/diffusion_helpers.cpp
families/ltx_video/runtime/diffusion_helpers.h
families/ltx_video/runtime/ltx_video_diffusion_types.h
families/ltx_video/runtime/ltx_video_scheduler_helpers.h
families/ltx_video/runtime/pipeline.cpp
families/ltx_video/runtime/pipeline.h
families/ltx_video/runtime/plugin.cpp
families/ltx_video/runtime/plugin_helpers.cpp
families/ltx_video/runtime/plugin_helpers.h
families/ltx_video/runtime/tokenizer.h
families/ltx_video/runtime/unigram_tokenizer.cpp
families/ltx_video/support.py
families/ltx_video/t5_encoder_builder.py
families/ltx_video/tests/__init__.py
families/ltx_video/tests/manifests/ltx-video-l0.json
families/ltx_video/tests/test_e2e.py
families/ltx_video/tests/thresholds/ltx-video-l0.json
families/m2m_100/__init__.py
families/m2m_100/checkpoint_mapper.py
families/m2m_100/config.py
families/m2m_100/graph_blocks.py
families/m2m_100/graph_ops.py
families/m2m_100/model.py
families/m2m_100/requirements.txt
families/m2m_100/runtime/CMakeLists.txt
families/m2m_100/runtime/decode_runtime.cpp
families/m2m_100/runtime/decode_runtime.h
families/m2m_100/runtime/inference_state.h
families/m2m_100/runtime/kv_cache.cpp
families/m2m_100/runtime/kv_cache.h
families/m2m_100/runtime/plugin.cpp
families/m2m_100/runtime/plugin_helpers.cpp
families/m2m_100/runtime/plugin_helpers.h
families/m2m_100/runtime/request_tokens.h
families/m2m_100/runtime/runtime_config.h
families/m2m_100/runtime/tokenizer.h
families/m2m_100/runtime/unigram_tokenizer.cpp
families/m2m_100/support.py
families/m2m_100/tests/__init__.py
families/m2m_100/tests/cpp/test_m2m_100_runtime_config.cpp
families/m2m_100/tests/cpp/test_m2m_100_tokenizer_schema.cpp
families/m2m_100/tests/manifests/nllb-200.json
families/m2m_100/tests/test_e2e.py
families/m2m_100/tests/thresholds/nllb-200-distilled-600m.json
families/magpie_tts/__init__.py
families/magpie_tts/checkpoint_mapper.py
families/magpie_tts/config.py
families/magpie_tts/decoder_tp_builder.py
families/magpie_tts/graph_ops.py
families/magpie_tts/magpie_tokenizer.py
families/magpie_tts/model.py
families/magpie_tts/nanocodec_builder.py
families/magpie_tts/parallel.py
families/magpie_tts/requirements.txt
families/magpie_tts/runtime/CMakeLists.txt
families/magpie_tts/runtime/audio_helpers.cpp
families/magpie_tts/runtime/audio_helpers.h
families/magpie_tts/runtime/cuda_common.cpp
families/magpie_tts/runtime/cuda_common.h
families/magpie_tts/runtime/decode_runtime.cpp
families/magpie_tts/runtime/decode_runtime.h
families/magpie_tts/runtime/inference_state.h
families/magpie_tts/runtime/ipa_tokenizer.cpp
families/magpie_tts/runtime/kv_cache.cpp
families/magpie_tts/runtime/kv_cache.h
families/magpie_tts/runtime/magpie_codec_plan.h
families/magpie_tts/runtime/magpie_config.h
families/magpie_tts/runtime/magpie_decode_policy.h
families/magpie_tts/runtime/magpie_decoder_plan.h
families/magpie_tts/runtime/magpie_generation_plan.h
families/magpie_tts/runtime/magpie_kernels.cu
families/magpie_tts/runtime/magpie_kernels.h
families/magpie_tts/runtime/magpie_text_completion_policy.h
families/magpie_tts/runtime/pipeline.cpp
families/magpie_tts/runtime/pipeline.h
families/magpie_tts/runtime/plugin.cpp
families/magpie_tts/runtime/plugin_helpers.cpp
families/magpie_tts/runtime/plugin_helpers.h
families/magpie_tts/runtime/tensor_names.h
families/magpie_tts/runtime/tokenizer.h
families/magpie_tts/support.py
families/magpie_tts/tests/__init__.py
families/magpie_tts/tests/manifests/magpie-tts-357m-tp4.json
families/magpie_tts/tests/manifests/magpie-tts-357m.json
families/magpie_tts/tests/test_e2e.py
families/magpie_tts/tests/thresholds/magpie-tts-357m-tp4.json
families/magpie_tts/tests/thresholds/magpie-tts-357m-tts-probe01.json
families/magpie_tts/tests/thresholds/magpie-tts-357m-tts-probe02.json
families/magpie_tts/tests/thresholds/magpie-tts-357m-tts-probe03.json
families/magpie_tts/tests/thresholds/magpie-tts-357m-tts-probe04.json
families/magpie_tts/tests/thresholds/magpie-tts-357m.json
families/mamba/__init__.py
families/mamba/checkpoint_mapper.py
families/mamba/config.py
families/mamba/graph_ops.py
families/mamba/model.py
families/mamba/parallel.py
families/mamba/runtime/CMakeLists.txt
families/mamba/runtime/bpe_tokenizer.cpp
families/mamba/runtime/chat_templates.cpp
families/mamba/runtime/chat_templates.h
families/mamba/runtime/distributed_runtime.cpp
families/mamba/runtime/distributed_runtime.h
families/mamba/runtime/inference_state.h
families/mamba/runtime/mamba_recurrent_step_contracts.h
families/mamba/runtime/pipeline.cpp
families/mamba/runtime/pipeline.h
families/mamba/runtime/plugin.cpp
families/mamba/runtime/plugin_helpers.cpp
families/mamba/runtime/plugin_helpers.h
families/mamba/runtime/recurrent_output_initializers.h
families/mamba/runtime/recurrent_state.cpp
families/mamba/runtime/recurrent_state.h
families/mamba/runtime/sampler.cpp
families/mamba/runtime/sampler.h
families/mamba/runtime/tokenizer.h
families/mamba/support.py
families/mamba/tests/__init__.py
families/mamba/tests/cpp/test_mamba_recurrent_output_initializers.cpp
families/mamba/tests/manifests/mamba-130m-tp4.json
families/mamba/tests/manifests/mamba-130m.json
families/mamba/tests/perf_validation.json
families/mamba/tests/test_e2e.py
families/mamba/tests/thresholds/mamba-130m-tp4.json
families/mamba/tests/thresholds/mamba-130m.json
families/mamba/tp_builder.py
families/marian/__init__.py
families/marian/checkpoint_mapper.py
families/marian/config.py
families/marian/decoder_tp_builder.py
families/marian/graph_ops.py
families/marian/model.py
families/marian/parallel.py
families/marian/runtime/CMakeLists.txt
families/marian/runtime/distributed_runtime.cpp
families/marian/runtime/distributed_runtime.h
families/marian/runtime/inference_state.h
families/marian/runtime/kv_cache.cpp
families/marian/runtime/kv_cache.h
families/marian/runtime/plugin.cpp
families/marian/runtime/plugin_helpers.cpp
families/marian/runtime/plugin_helpers.h
families/marian/runtime/runtime_config.h
families/marian/runtime/tokenizer.h
families/marian/runtime/unigram_tokenizer.cpp
families/marian/support.py
families/marian/tests/__init__.py
families/marian/tests/cpp/test_marian_runtime_config.cpp
families/marian/tests/manifests/marian-en-ru-tp4.json
families/marian/tests/manifests/marian-en-ru.json
families/marian/tests/test_e2e.py
families/marian/tests/thresholds/marian-en-ru-tp4.json
families/marian/tests/thresholds/marian-en-ru.json
families/marian/tokenizer_conversion.py
families/marian/tokenizer_json.py
families/minimax_h3/__init__.py
families/minimax_h3/adaln_builder.py
families/minimax_h3/checkpoint.py
families/minimax_h3/config.py
families/minimax_h3/dit_builder.py
families/minimax_h3/graph_ops.py
families/minimax_h3/model.py
families/minimax_h3/requirements.txt
families/minimax_h3/runtime/CMakeLists.txt
families/minimax_h3/runtime/bpe_tokenizer.cpp
families/minimax_h3/runtime/pipeline.cpp
families/minimax_h3/runtime/pipeline.h
families/minimax_h3/runtime/plugin.cpp
families/minimax_h3/runtime/tokenizer.h
families/minimax_h3/runtime/torch_cuda_normal.cu
families/minimax_h3/runtime/torch_cuda_normal.h
families/minimax_h3/support.py
families/minimax_h3/tests/__init__.py
families/minimax_h3/tests/cpp/test_minimax_h3_math.cpp
families/minimax_h3/tests/manifests/minimax-h3-768p.json
families/minimax_h3/tests/prompts/t2va-example-1.json
families/minimax_h3/tests/test_config.py
families/minimax_h3/tests/test_dynamic_shapes.py
families/minimax_h3/tests/test_e2e.py
families/minimax_h3/tests/test_trt_builders.py
families/minimax_h3/tests/thresholds/minimax-h3-768p.json
families/minimax_h3/text_encoder_builder.py
families/minimax_h3/vae_builder.py
families/mistral/__init__.py
families/mistral/checkpoint_mapper.py
families/mistral/config.py
families/mistral/default_decoder.py
families/mistral/default_dual_profile_decoder.py
families/mistral/graph_blocks.py
families/mistral/graph_ops.py
families/mistral/model.py
families/mistral/runtime/CMakeLists.txt
families/mistral/runtime/bpe_tokenizer.cpp
families/mistral/runtime/chat_templates.cpp
families/mistral/runtime/chat_templates.h
families/mistral/runtime/inference_state.h
families/mistral/runtime/kv_cache.cpp
families/mistral/runtime/kv_cache.h
families/mistral/runtime/pipeline.cpp
families/mistral/runtime/pipeline.h
families/mistral/runtime/plugin.cpp
families/mistral/runtime/plugin_helpers.cpp
families/mistral/runtime/plugin_helpers.h
families/mistral/runtime/sampler.cpp
families/mistral/runtime/sampler.h
families/mistral/runtime/tensor_names.h
families/mistral/runtime/tokenizer.h
families/mistral/support.py
families/mistral/tests/__init__.py
families/mistral/tests/manifests/mistral-7b-l0.json
families/mistral/tests/manifests/mistral-7b.json
families/mistral/tests/manifests/riva-translate-4b.json
families/mistral/tests/test_e2e.py
families/mistral/tests/thresholds/mistral-7b-l0.json
families/mistral/tests/thresholds/mistral-7b.json
families/mistral/tests/thresholds/riva-translate-4b.json
families/mistral/utils.py
families/mixtral/__init__.py
families/mixtral/checkpoint_mapper.py
families/mixtral/config.py
families/mixtral/default_decoder.py
families/mixtral/default_dual_profile_decoder.py
families/mixtral/graph_blocks.py
families/mixtral/graph_ops.py
families/mixtral/model.py
families/mixtral/parallel.py
families/mixtral/runtime/CMakeLists.txt
families/mixtral/runtime/bpe_tokenizer.cpp
families/mixtral/runtime/chat_templates.cpp
families/mixtral/runtime/chat_templates.h
families/mixtral/runtime/distributed_runtime.cpp
families/mixtral/runtime/distributed_runtime.h
families/mixtral/runtime/inference_state.h
families/mixtral/runtime/kv_cache.cpp
families/mixtral/runtime/kv_cache.h
families/mixtral/runtime/pipeline.cpp
families/mixtral/runtime/pipeline.h
families/mixtral/runtime/plugin.cpp
families/mixtral/runtime/plugin_helpers.cpp
families/mixtral/runtime/plugin_helpers.h
families/mixtral/runtime/sampler.cpp
families/mixtral/runtime/sampler.h
families/mixtral/runtime/tensor_names.h
families/mixtral/runtime/tokenizer.h
families/mixtral/support.py
families/mixtral/tests/__init__.py
families/mixtral/tests/manifests/mixtral-stories-15m-tp2.json
families/mixtral/tests/manifests/mixtral-stories-15m.json
families/mixtral/tests/perf_validation.json
families/mixtral/tests/test_e2e.py
families/mixtral/tests/thresholds/mixtral-stories-15m-tp2.json
families/mixtral/tests/thresholds/mixtral-stories-15m.json
families/mixtral/tp_builder.py
families/mixtral/utils.py
families/modernbert/__init__.py
families/modernbert/config.py
families/modernbert/graph/__init__.py
families/modernbert/graph/model.py
families/modernbert/model.py
families/modernbert/parallel.py
families/modernbert/runtime/CMakeLists.txt
families/modernbert/runtime/bpe_tokenizer.cpp
families/modernbert/runtime/pipeline.cpp
families/modernbert/runtime/pipeline.h
families/modernbert/runtime/plugin.cpp
families/modernbert/runtime/plugin_helpers.cpp
families/modernbert/runtime/plugin_helpers.h
families/modernbert/runtime/tokenizer.h
families/modernbert/support.py
families/modernbert/tests/__init__.py
families/modernbert/tests/manifests/modernbert-base-tp4.json
families/modernbert/tests/manifests/modernbert-base.json
families/modernbert/tests/test_e2e.py
families/modernbert/tests/thresholds/modernbert-base-tp4.json
families/modernbert/tests/thresholds/modernbert-base.json
families/modernbert/weights/__init__.py
families/moge/__init__.py
families/moge/model.py
families/moge/requirements.txt
families/moge/runtime/CMakeLists.txt
families/moge/runtime/pipeline.cpp
families/moge/runtime/pipeline.h
families/moge/runtime/plugin.cpp
families/moge/support.py
families/moge/tests/__init__.py
families/moge/tests/cpp/test_moge_pipeline.cpp
families/moge/tests/manifests/moge-2-vitl.json
families/moge/tests/official_reference.py
families/moge/tests/reference-source.json
families/moge/tests/test_e2e.py
families/moge/tests/test_fused_resample.py
families/moge/tests/test_selective_zero_padding.py
families/moge/tests/thresholds/moge-2-vitl.json
families/mpnet/__init__.py
families/mpnet/checkpoint_mapper.py
families/mpnet/config.py
families/mpnet/encoder_builder.py
families/mpnet/graph_ops.py
families/mpnet/model.py
families/mpnet/parallel.py
families/mpnet/runtime/CMakeLists.txt
families/mpnet/runtime/pipeline.cpp
families/mpnet/runtime/pipeline.h
families/mpnet/runtime/plugin.cpp
families/mpnet/runtime/plugin_helpers.cpp
families/mpnet/runtime/plugin_helpers.h
families/mpnet/runtime/tokenizer.h
families/mpnet/runtime/wordpiece_tokenizer.cpp
families/mpnet/support.py
families/mpnet/tests/__init__.py
families/mpnet/tests/manifests/all-mpnet-base-v2-tp4.json
families/mpnet/tests/manifests/all-mpnet-base-v2.json
families/mpnet/tests/test_e2e.py
families/mpnet/tests/thresholds/all-mpnet-base-v2-tp4.json
families/mpnet/tests/thresholds/all-mpnet-base-v2.json
families/mpnet/tp_builder.py
families/nemotron/__init__.py
families/nemotron/checkpoint_mapper.py
families/nemotron/config.py
families/nemotron/default_decoder.py
families/nemotron/default_dual_profile_decoder.py
families/nemotron/graph_blocks.py
families/nemotron/graph_ops.py
families/nemotron/model.py
families/nemotron/runtime/CMakeLists.txt
families/nemotron/runtime/bpe_tokenizer.cpp
families/nemotron/runtime/chat_templates.cpp
families/nemotron/runtime/chat_templates.h
families/nemotron/runtime/inference_state.h
families/nemotron/runtime/kv_cache.cpp
families/nemotron/runtime/kv_cache.h
families/nemotron/runtime/pipeline.cpp
families/nemotron/runtime/pipeline.h
families/nemotron/runtime/plugin.cpp
families/nemotron/runtime/plugin_helpers.cpp
families/nemotron/runtime/plugin_helpers.h
families/nemotron/runtime/sampler.cpp
families/nemotron/runtime/sampler.h
families/nemotron/runtime/tensor_names.h
families/nemotron/runtime/tokenizer.h
families/nemotron/support.py
families/nemotron/tests/__init__.py
families/nemotron/tests/manifests/nemotron-hindi-4b.json
families/nemotron/tests/manifests/nemotron-mini-4b.json
families/nemotron/tests/test_chat_template_contract.py
families/nemotron/tests/test_e2e.py
families/nemotron/tests/thresholds/nemotron-hindi-4b.json
families/nemotron/tests/thresholds/nemotron-mini-4b.json
families/nemotron/utils.py
families/nemotron_h/__init__.py
families/nemotron_h/checkpoint_mapper.py
families/nemotron_h/config.py
families/nemotron_h/graph_blocks.py
families/nemotron_h/graph_ops.py
families/nemotron_h/model.py
families/nemotron_h/parallel.py
families/nemotron_h/requirements.txt
families/nemotron_h/runtime/CMakeLists.txt
families/nemotron_h/runtime/bpe_tokenizer.cpp
families/nemotron_h/runtime/chat_templates.cpp
families/nemotron_h/runtime/chat_templates.h
families/nemotron_h/runtime/distributed_runtime.cpp
families/nemotron_h/runtime/distributed_runtime.h
families/nemotron_h/runtime/hybrid_state.cpp
families/nemotron_h/runtime/hybrid_state.h
families/nemotron_h/runtime/inference_state.h
families/nemotron_h/runtime/kv_cache.cpp
families/nemotron_h/runtime/kv_cache.h
families/nemotron_h/runtime/nemotron_h_recurrent_step_contracts.h
families/nemotron_h/runtime/pipeline.cpp
families/nemotron_h/runtime/pipeline.h
families/nemotron_h/runtime/plugin.cpp
families/nemotron_h/runtime/plugin_helpers.cpp
families/nemotron_h/runtime/plugin_helpers.h
families/nemotron_h/runtime/recurrent_generation_plan.h
families/nemotron_h/runtime/recurrent_output_initializers.h
families/nemotron_h/runtime/recurrent_state.cpp
families/nemotron_h/runtime/recurrent_state.h
families/nemotron_h/runtime/sampler.cpp
families/nemotron_h/runtime/sampler.h
families/nemotron_h/runtime/tokenizer.h
families/nemotron_h/support.py
families/nemotron_h/tests/__init__.py
families/nemotron_h/tests/cpp/test_nemotron_h_recurrent_output_initializers.cpp
families/nemotron_h/tests/data/nemotron_h_nano_9b_golden.json
families/nemotron_h/tests/manifests/nemotron-h-nano-9b-tp4.json
families/nemotron_h/tests/manifests/nemotron-h-nano-9b.json
families/nemotron_h/tests/test_e2e.py
families/nemotron_h/tests/test_runtime_contract.py
families/nemotron_h/tests/thresholds/nemotron-h-nano-9b-tp4.json
families/nemotron_h/tests/thresholds/nemotron-h-nano-9b.json
families/nemotron_h/tp_builder.py
families/nemotron_labs_diffusion/__init__.py
families/nemotron_labs_diffusion/checkpoint_mapper.py
families/nemotron_labs_diffusion/config.py
families/nemotron_labs_diffusion/default_decoder.py
families/nemotron_labs_diffusion/default_dual_profile_decoder.py
families/nemotron_labs_diffusion/graph_blocks.py
families/nemotron_labs_diffusion/graph_ops.py
families/nemotron_labs_diffusion/model.py
families/nemotron_labs_diffusion/requirements.txt
families/nemotron_labs_diffusion/runtime/CMakeLists.txt
families/nemotron_labs_diffusion/runtime/bpe_tokenizer.cpp
families/nemotron_labs_diffusion/runtime/chat_templates.cpp
families/nemotron_labs_diffusion/runtime/chat_templates.h
families/nemotron_labs_diffusion/runtime/inference_state.h
families/nemotron_labs_diffusion/runtime/kv_cache.cpp
families/nemotron_labs_diffusion/runtime/kv_cache.h
families/nemotron_labs_diffusion/runtime/pipeline.cpp
families/nemotron_labs_diffusion/runtime/pipeline.h
families/nemotron_labs_diffusion/runtime/plugin.cpp
families/nemotron_labs_diffusion/runtime/plugin_helpers.cpp
families/nemotron_labs_diffusion/runtime/plugin_helpers.h
families/nemotron_labs_diffusion/runtime/runtime_config.h
families/nemotron_labs_diffusion/runtime/sampler.cpp
families/nemotron_labs_diffusion/runtime/sampler.h
families/nemotron_labs_diffusion/runtime/tensor_names.h
families/nemotron_labs_diffusion/runtime/tokenizer.h
families/nemotron_labs_diffusion/support.py
families/nemotron_labs_diffusion/tests/__init__.py
families/nemotron_labs_diffusion/tests/cpp/test_nemotron_labs_diffusion_runtime_config.cpp
families/nemotron_labs_diffusion/tests/manifests/nemotron-labs-diffusion-8b-l0.json
families/nemotron_labs_diffusion/tests/manifests/nemotron-labs-diffusion-8b.json
families/nemotron_labs_diffusion/tests/test_e2e.py
families/nemotron_labs_diffusion/tests/thresholds/nemotron-labs-diffusion-8b-ar.json
families/nemotron_labs_diffusion/tests/thresholds/nemotron-labs-diffusion-8b-diffusion.json
families/nemotron_labs_diffusion/tests/thresholds/nemotron-labs-diffusion-8b-l0.json
families/nemotron_labs_diffusion/tests/thresholds/nemotron-labs-diffusion-8b-linear-spec-lora.json
families/nemotron_labs_diffusion/tests/thresholds/nemotron-labs-diffusion-8b-linear-spec.json
families/nemotron_labs_diffusion/utils.py
families/nemotron_speech_streaming/__init__.py
families/nemotron_speech_streaming/canary_encoder_helpers.py
families/nemotron_speech_streaming/checkpoint_mapper.py
families/nemotron_speech_streaming/config.py
families/nemotron_speech_streaming/graph_ops.py
families/nemotron_speech_streaming/model.py
families/nemotron_speech_streaming/parallel.py
families/nemotron_speech_streaming/predictor_tp_builder.py
families/nemotron_speech_streaming/requirements.txt
families/nemotron_speech_streaming/runtime/CMakeLists.txt
families/nemotron_speech_streaming/runtime/audio_helpers.cpp
families/nemotron_speech_streaming/runtime/audio_helpers.h
families/nemotron_speech_streaming/runtime/pipeline.cpp
families/nemotron_speech_streaming/runtime/pipeline.h
families/nemotron_speech_streaming/runtime/plugin.cpp
families/nemotron_speech_streaming/runtime/plugin_helpers.cpp
families/nemotron_speech_streaming/runtime/plugin_helpers.h
families/nemotron_speech_streaming/runtime/resampler.cpp
families/nemotron_speech_streaming/runtime/resampler.h
families/nemotron_speech_streaming/runtime/rnnt_config.h
families/nemotron_speech_streaming/runtime/tokenizer.h
families/nemotron_speech_streaming/runtime/unigram_tokenizer.cpp
families/nemotron_speech_streaming/support.py
families/nemotron_speech_streaming/tests/__init__.py
families/nemotron_speech_streaming/tests/data/asr_probes/README.md
families/nemotron_speech_streaming/tests/data/asr_probes/generate_asr_probe_inputs.py
families/nemotron_speech_streaming/tests/data/asr_probes/manifest.json
families/nemotron_speech_streaming/tests/manifests/nemotron-3.5-asr-streaming-0.6b.json
families/nemotron_speech_streaming/tests/manifests/nemotron-speech-streaming-en-0.6b-tp4.json
families/nemotron_speech_streaming/tests/manifests/nemotron-speech-streaming-en-0.6b.json
families/nemotron_speech_streaming/tests/test_e2e.py
families/nemotron_speech_streaming/tests/thresholds/nemotron-3.5-asr-streaming-0.6b.json
families/nemotron_speech_streaming/tests/thresholds/nemotron-speech-streaming-en-0.6b-asr-probe01.json
families/nemotron_speech_streaming/tests/thresholds/nemotron-speech-streaming-en-0.6b-asr-probe02.json
families/nemotron_speech_streaming/tests/thresholds/nemotron-speech-streaming-en-0.6b-asr-probe03.json
families/nemotron_speech_streaming/tests/thresholds/nemotron-speech-streaming-en-0.6b-asr-probe04.json
families/nemotron_speech_streaming/tests/thresholds/nemotron-speech-streaming-en-0.6b-asr-probe05.json
families/nemotron_speech_streaming/tests/thresholds/nemotron-speech-streaming-en-0.6b-asr-probe06.json
families/nemotron_speech_streaming/tests/thresholds/nemotron-speech-streaming-en-0.6b-asr-probe08.json
families/nemotron_speech_streaming/tests/thresholds/nemotron-speech-streaming-en-0.6b-tp4.json
families/nemotron_speech_streaming/tests/thresholds/nemotron-speech-streaming-en-0.6b.json
families/nemotron_voicechat/__init__.py
families/nemotron_voicechat/checkpoint_mapper.py
families/nemotron_voicechat/config.py
families/nemotron_voicechat/conformer.py
families/nemotron_voicechat/graph_blocks.py
families/nemotron_voicechat/graph_ops.py
families/nemotron_voicechat/model.py
families/nemotron_voicechat/native_codec.py
families/nemotron_voicechat/native_core.py
families/nemotron_voicechat/native_tts.py
families/nemotron_voicechat/requirements.txt
families/nemotron_voicechat/runtime/CMakeLists.txt
families/nemotron_voicechat/runtime/audio_helpers.cpp
families/nemotron_voicechat/runtime/audio_helpers.h
families/nemotron_voicechat/runtime/bpe_tokenizer.cpp
families/nemotron_voicechat/runtime/codec_reconstruction.cpp
families/nemotron_voicechat/runtime/codec_reconstruction.h
families/nemotron_voicechat/runtime/function_channel.cpp
families/nemotron_voicechat/runtime/function_channel.h
families/nemotron_voicechat/runtime/pipeline.cpp
families/nemotron_voicechat/runtime/pipeline.h
families/nemotron_voicechat/runtime/plugin.cpp
families/nemotron_voicechat/runtime/plugin_helpers.cpp
families/nemotron_voicechat/runtime/plugin_helpers.h
families/nemotron_voicechat/runtime/resampler.cpp
families/nemotron_voicechat/runtime/resampler.h
families/nemotron_voicechat/runtime/session_state.cpp
families/nemotron_voicechat/runtime/session_state.h
families/nemotron_voicechat/runtime/thinker_hybrid_state.cpp
families/nemotron_voicechat/runtime/thinker_hybrid_state.h
families/nemotron_voicechat/runtime/thinker_inference_state.h
families/nemotron_voicechat/runtime/thinker_kv_cache.cpp
families/nemotron_voicechat/runtime/thinker_kv_cache.h
families/nemotron_voicechat/runtime/thinker_mamba_state.cpp
families/nemotron_voicechat/runtime/thinker_mamba_state.h
families/nemotron_voicechat/runtime/tokenizer.h
families/nemotron_voicechat/runtime/voicechat_config.h
families/nemotron_voicechat/streaming_perception.py
families/nemotron_voicechat/support.py
families/nemotron_voicechat/tests/__init__.py
families/nemotron_voicechat/tests/cpp/native_lifecycle_probe.cpp
families/nemotron_voicechat/tests/lifecycle_oracle.py
families/nemotron_voicechat/tests/manifests/nemotron-voicechat-11b.json
families/nemotron_voicechat/tests/reference-source.json
families/nemotron_voicechat/tests/test_e2e.py
families/nemotron_voicechat/tests/test_lifecycle_oracle.py
families/nemotron_voicechat/tests/test_model_card_oracle.py
families/nemotron_voicechat/tests/thresholds/nemotron-voicechat-11b-full-duplex-lifecycle.json
families/nemotron_voicechat/tests/thresholds/nemotron-voicechat-11b.json
families/olmo/__init__.py
families/olmo/checkpoint_mapper.py
families/olmo/config.py
families/olmo/default_decoder.py
families/olmo/default_dual_profile_decoder.py
families/olmo/dual_profile_decoder_tp_builder.py
families/olmo/graph_blocks.py
families/olmo/graph_ops.py
families/olmo/model.py
families/olmo/parallel.py
families/olmo/runtime/CMakeLists.txt
families/olmo/runtime/bpe_tokenizer.cpp
families/olmo/runtime/chat_templates.cpp
families/olmo/runtime/chat_templates.h
families/olmo/runtime/distributed_runtime.cpp
families/olmo/runtime/distributed_runtime.h
families/olmo/runtime/inference_state.h
families/olmo/runtime/kv_cache.cpp
families/olmo/runtime/kv_cache.h
families/olmo/runtime/pipeline.cpp
families/olmo/runtime/pipeline.h
families/olmo/runtime/plugin.cpp
families/olmo/runtime/plugin_helpers.cpp
families/olmo/runtime/plugin_helpers.h
families/olmo/runtime/sampler.cpp
families/olmo/runtime/sampler.h
families/olmo/runtime/tensor_names.h
families/olmo/runtime/tokenizer.h
families/olmo/support.py
families/olmo/tests/__init__.py
families/olmo/tests/manifests/olmo-1b-tp4.json
families/olmo/tests/manifests/olmo-1b.json
families/olmo/tests/test_e2e.py
families/olmo/tests/thresholds/olmo-1b-tp4.json
families/olmo/tests/thresholds/olmo-1b.json
families/olmo/utils.py
families/olmo2/__init__.py
families/olmo2/checkpoint_mapper.py
families/olmo2/config.py
families/olmo2/graph_blocks.py
families/olmo2/graph_ops.py
families/olmo2/model.py
families/olmo2/parallel.py
families/olmo2/prefill_builder.py
families/olmo2/runtime/CMakeLists.txt
families/olmo2/runtime/bpe_tokenizer.cpp
families/olmo2/runtime/chat_templates.cpp
families/olmo2/runtime/chat_templates.h
families/olmo2/runtime/distributed_runtime.cpp
families/olmo2/runtime/distributed_runtime.h
families/olmo2/runtime/inference_state.h
families/olmo2/runtime/kv_cache.cpp
families/olmo2/runtime/kv_cache.h
families/olmo2/runtime/pipeline.cpp
families/olmo2/runtime/pipeline.h
families/olmo2/runtime/plugin.cpp
families/olmo2/runtime/plugin_helpers.cpp
families/olmo2/runtime/plugin_helpers.h
families/olmo2/runtime/sampler.cpp
families/olmo2/runtime/sampler.h
families/olmo2/runtime/tensor_names.h
families/olmo2/runtime/tokenizer.h
families/olmo2/support.py
families/olmo2/tests/__init__.py
families/olmo2/tests/manifests/olmo2-1b-tp4.json
families/olmo2/tests/manifests/olmo2-1b.json
families/olmo2/tests/test_e2e.py
families/olmo2/tests/thresholds/olmo2-1b-tp4.json
families/olmo2/tests/thresholds/olmo2-1b.json
families/olmo2/tp_builder.py
families/opt/__init__.py
families/opt/checkpoint_mapper.py
families/opt/config.py
families/opt/default_decoder.py
families/opt/default_dual_profile_decoder.py
families/opt/default_dual_profile_decoder_tp.py
families/opt/graph_blocks.py
families/opt/graph_ops.py
families/opt/model.py
families/opt/parallel.py
families/opt/runtime/CMakeLists.txt
families/opt/runtime/bpe_tokenizer.cpp
families/opt/runtime/chat_templates.cpp
families/opt/runtime/chat_templates.h
families/opt/runtime/distributed_runtime.cpp
families/opt/runtime/distributed_runtime.h
families/opt/runtime/inference_state.h
families/opt/runtime/kv_cache.cpp
families/opt/runtime/kv_cache.h
families/opt/runtime/pipeline.cpp
families/opt/runtime/pipeline.h
families/opt/runtime/plugin.cpp
families/opt/runtime/plugin_helpers.cpp
families/opt/runtime/plugin_helpers.h
families/opt/runtime/sampler.cpp
families/opt/runtime/sampler.h
families/opt/runtime/tensor_names.h
families/opt/runtime/tokenizer.h
families/opt/support.py
families/opt/tests/__init__.py
families/opt/tests/manifests/opt-125m-tp4.json
families/opt/tests/manifests/opt-125m.json
families/opt/tests/test_e2e.py
families/opt/tests/thresholds/opt-125m-tp4.json
families/opt/tests/thresholds/opt-125m.json
families/opt/utils.py
families/patchtsmixer/__init__.py
families/patchtsmixer/checkpoint_mapper.py
families/patchtsmixer/graph_ops.py
families/patchtsmixer/model.py
families/patchtsmixer/runtime/CMakeLists.txt
families/patchtsmixer/runtime/pipeline.cpp
families/patchtsmixer/runtime/pipeline.h
families/patchtsmixer/runtime/plugin.cpp
families/patchtsmixer/runtime/plugin_helpers.cpp
families/patchtsmixer/runtime/plugin_helpers.h
families/patchtsmixer/support.py
families/patchtsmixer/tests/__init__.py
families/patchtsmixer/tests/cpp/test_forecast_contract.cpp
families/patchtsmixer/tests/etth1.py
families/patchtsmixer/tests/manifests/patchtsmixer-granite-official-tp4.json
families/patchtsmixer/tests/manifests/patchtsmixer-granite-official.json
families/patchtsmixer/tests/reference-source.json
families/patchtsmixer/tests/test_e2e.py
families/patchtsmixer/tests/test_etth1.py
families/patchtsmixer/tests/thresholds/patchtsmixer-granite-official-tp4.json
families/patchtsmixer/tests/thresholds/patchtsmixer-granite-official.json
families/patchtsmixer/time_series_trt.py
families/patchtst/__init__.py
families/patchtst/checkpoint_mapper.py
families/patchtst/config.py
families/patchtst/graph_ops.py
families/patchtst/model.py
families/patchtst/runtime/CMakeLists.txt
families/patchtst/runtime/pipeline.cpp
families/patchtst/runtime/pipeline.h
families/patchtst/runtime/plugin.cpp
families/patchtst/support.py
families/patchtst/tests/__init__.py
families/patchtst/tests/cpp/test_forecast_contract.cpp
families/patchtst/tests/etth1.py
families/patchtst/tests/manifests/patchtst-etth1-regression-distribution-tp4.json
families/patchtst/tests/manifests/patchtst-etth1-regression-distribution.json
families/patchtst/tests/manifests/patchtst-granite-official.json
families/patchtst/tests/reference-source.json
families/patchtst/tests/test_e2e.py
families/patchtst/tests/test_etth1.py
families/patchtst/tests/thresholds/patchtst-etth1-regression-distribution-tp4.json
families/patchtst/tests/thresholds/patchtst-etth1-regression-distribution.json
families/patchtst/tests/thresholds/patchtst-granite-official.json
families/patchtst/time_series_trt.py
families/personaplex/__init__.py
families/personaplex/checkpoint_mapper.py
families/personaplex/config.py
families/personaplex/decoder_tp_builder.py
families/personaplex/default_decoder.py
families/personaplex/default_dual_profile_decoder.py
families/personaplex/graph_blocks.py
families/personaplex/graph_ops.py
families/personaplex/mimi_streaming_encoder.py
families/personaplex/mimi_weights.py
families/personaplex/model.py
families/personaplex/parallel.py
families/personaplex/requirements.txt
families/personaplex/runtime/CMakeLists.txt
families/personaplex/runtime/decode_runtime.cpp
families/personaplex/runtime/decode_runtime.h
families/personaplex/runtime/inference_state.h
families/personaplex/runtime/kv_cache.cpp
families/personaplex/runtime/kv_cache.h
families/personaplex/runtime/pipeline.cpp
families/personaplex/runtime/pipeline.h
families/personaplex/runtime/plugin.cpp
families/personaplex/runtime/plugin_helpers.cpp
families/personaplex/runtime/plugin_helpers.h
families/personaplex/runtime/resampler.cpp
families/personaplex/runtime/resampler.h
families/personaplex/runtime/sampling_kernels.cu
families/personaplex/runtime/sampling_kernels.h
families/personaplex/runtime/speech_config.h
families/personaplex/runtime/speech_decode_stop_policy.h
families/personaplex/runtime/speech_delay_cache.h
families/personaplex/runtime/speech_depth_plan.h
families/personaplex/runtime/speech_generation_policy.h
families/personaplex/runtime/speech_mimi_decode_plan.h
families/personaplex/runtime/speech_mimi_encode_plan.h
families/personaplex/runtime/speech_performance.h
families/personaplex/runtime/speech_runtime_plan.h
families/personaplex/runtime/speech_temporal_embed_plan.h
families/personaplex/runtime/speech_waveform_postprocess.h
families/personaplex/support.py
families/personaplex/tests/__init__.py
families/personaplex/tests/data/personaplex_recording_official_tokens_greedy.npy
families/personaplex/tests/manifests/personaplex-7b-l0-tp4.json
families/personaplex/tests/manifests/personaplex-7b-l0.json
families/personaplex/tests/manifests/personaplex-7b.json
families/personaplex/tests/test_e2e.py
families/personaplex/tests/thresholds/personaplex-7b-l0-tp4.json
families/personaplex/tests/thresholds/personaplex-7b-l0.json
families/personaplex/tests/thresholds/personaplex-7b.json
families/personaplex/utils.py
families/phi/__init__.py
families/phi/checkpoint_mapper.py
families/phi/config.py
families/phi/default_decoder.py
families/phi/default_dual_profile_decoder.py
families/phi/dual_profile_decoder_tp_builder.py
families/phi/graph_blocks.py
families/phi/graph_ops.py
families/phi/model.py
families/phi/parallel.py
families/phi/runtime/CMakeLists.txt
families/phi/runtime/bpe_tokenizer.cpp
families/phi/runtime/chat_templates.cpp
families/phi/runtime/chat_templates.h
families/phi/runtime/distributed_runtime.cpp
families/phi/runtime/distributed_runtime.h
families/phi/runtime/inference_state.h
families/phi/runtime/kv_cache.cpp
families/phi/runtime/kv_cache.h
families/phi/runtime/pipeline.cpp
families/phi/runtime/pipeline.h
families/phi/runtime/plugin.cpp
families/phi/runtime/plugin_helpers.cpp
families/phi/runtime/plugin_helpers.h
families/phi/runtime/sampler.cpp
families/phi/runtime/sampler.h
families/phi/runtime/tensor_names.h
families/phi/runtime/tokenizer.h
families/phi/support.py
families/phi/tests/__init__.py
families/phi/tests/manifests/phi3-mini-tp4.json
families/phi/tests/manifests/phi3-mini.json
families/phi/tests/perf_validation.json
families/phi/tests/test_e2e.py
families/phi/tests/thresholds/phi3-mini-tp4.json
families/phi/tests/thresholds/phi3-mini.json
families/phi/utils.py
families/phi4_multimodal/__init__.py
families/phi4_multimodal/checkpoint_mapper.py
families/phi4_multimodal/config.py
families/phi4_multimodal/default_decoder.py
families/phi4_multimodal/default_dual_profile_decoder.py
families/phi4_multimodal/graph_blocks.py
families/phi4_multimodal/graph_ops.py
families/phi4_multimodal/model.py
families/phi4_multimodal/phi4mm_vision_builder.py
families/phi4_multimodal/requirements.txt
families/phi4_multimodal/runtime/CMakeLists.txt
families/phi4_multimodal/runtime/bpe_tokenizer.cpp
families/phi4_multimodal/runtime/cuda_stream.h
families/phi4_multimodal/runtime/decoded_image.h
families/phi4_multimodal/runtime/image_preprocessor.cpp
families/phi4_multimodal/runtime/image_preprocessor.h
families/phi4_multimodal/runtime/image_transform_helper.h
families/phi4_multimodal/runtime/inference_state.h
families/phi4_multimodal/runtime/kv_cache.cpp
families/phi4_multimodal/runtime/kv_cache.h
families/phi4_multimodal/runtime/pipeline.cpp
families/phi4_multimodal/runtime/pipeline.h
families/phi4_multimodal/runtime/plugin.cpp
families/phi4_multimodal/runtime/plugin_helpers.cpp
families/phi4_multimodal/runtime/plugin_helpers.h
families/phi4_multimodal/runtime/sampler.cpp
families/phi4_multimodal/runtime/sampler.h
families/phi4_multimodal/runtime/tensor_names.h
families/phi4_multimodal/runtime/tokenizer.h
families/phi4_multimodal/support.py
families/phi4_multimodal/tests/__init__.py
families/phi4_multimodal/tests/manifests/phi4-multimodal.json
families/phi4_multimodal/tests/test_checkpoint_contract.py
families/phi4_multimodal/tests/test_e2e.py
families/phi4_multimodal/tests/test_vision_oracle.py
families/phi4_multimodal/tests/thresholds/phi4-multimodal.json
families/phi4_multimodal/tests/vision_oracle.py
families/phi4_multimodal/utils.py
families/phi_moe/__init__.py
families/phi_moe/checkpoint_mapper.py
families/phi_moe/config.py
families/phi_moe/default_decoder.py
families/phi_moe/default_dual_profile_decoder.py
families/phi_moe/graph_blocks.py
families/phi_moe/graph_ops.py
families/phi_moe/model.py
families/phi_moe/parallel.py
families/phi_moe/runtime/CMakeLists.txt
families/phi_moe/runtime/bpe_tokenizer.cpp
families/phi_moe/runtime/chat_templates.cpp
families/phi_moe/runtime/chat_templates.h
families/phi_moe/runtime/distributed_runtime.cpp
families/phi_moe/runtime/distributed_runtime.h
families/phi_moe/runtime/inference_state.h
families/phi_moe/runtime/kv_cache.cpp
families/phi_moe/runtime/kv_cache.h
families/phi_moe/runtime/pipeline.cpp
families/phi_moe/runtime/pipeline.h
families/phi_moe/runtime/plugin.cpp
families/phi_moe/runtime/plugin_helpers.cpp
families/phi_moe/runtime/plugin_helpers.h
families/phi_moe/runtime/sampler.cpp
families/phi_moe/runtime/sampler.h
families/phi_moe/runtime/tensor_names.h
families/phi_moe/runtime/tokenizer.h
families/phi_moe/support.py
families/phi_moe/tests/__init__.py
families/phi_moe/tests/manifests/phi-moe-l0-tp4.json
families/phi_moe/tests/manifests/phi-moe-l0.json
families/phi_moe/tests/manifests/phi-moe.json
families/phi_moe/tests/test_chat_template_contract.py
families/phi_moe/tests/test_e2e.py
families/phi_moe/tests/thresholds/phi-moe-l0-tp4.json
families/phi_moe/tests/thresholds/phi-moe-l0.json
families/phi_moe/tests/thresholds/phi-moe.json
families/phi_moe/tp_builder.py
families/phi_moe/utils.py
families/pixart/__init__.py
families/pixart/checkpoint_mapper.py
families/pixart/config.py
families/pixart/graph_ops.py
families/pixart/model.py
families/pixart/parallel.py
families/pixart/requirements.txt
families/pixart/runtime/CMakeLists.txt
families/pixart/runtime/diffusion_helpers.cpp
families/pixart/runtime/diffusion_helpers.h
families/pixart/runtime/gpu_matmul.cpp
families/pixart/runtime/gpu_matmul.h
families/pixart/runtime/pipeline.cpp
families/pixart/runtime/pipeline.h
families/pixart/runtime/pixart_denoising_step_seam.h
families/pixart/runtime/pixart_diffusion_types.h
families/pixart/runtime/pixart_dpmsolver.h
families/pixart/runtime/pixart_generation_conditioning.h
families/pixart/runtime/pixart_generation_plan.h
families/pixart/runtime/pixart_matmul_policy.h
families/pixart/runtime/pixart_scheduler_helpers.h
families/pixart/runtime/plugin.cpp
families/pixart/runtime/plugin_helpers.cpp
families/pixart/runtime/plugin_helpers.h
families/pixart/runtime/preprocessor_weights_helpers.h
families/pixart/runtime/tokenizer.h
families/pixart/runtime/unigram_tokenizer.cpp
families/pixart/standard_dit_builder.py
families/pixart/standard_dit_tp_builder.py
families/pixart/support.py
families/pixart/t5_encoder_builder.py
families/pixart/tests/__init__.py
families/pixart/tests/manifests/pixart-sigma-1024-l0.json
families/pixart/tests/manifests/pixart-sigma-1024-tp4.json
families/pixart/tests/manifests/pixart-sigma-1024.json
families/pixart/tests/test_e2e.py
families/pixart/tests/thresholds/pixart-sigma-1024-l0.json
families/pixart/tests/thresholds/pixart-sigma-1024-tp4.json
families/pixart/tests/thresholds/pixart-sigma-1024.json
families/pixart/vae_2d_builder.py
families/qwen/__init__.py
families/qwen/build_routing.py
families/qwen/checkpoint_mapper.py
families/qwen/config.py
families/qwen/dual_profile_decoder_builder.py
families/qwen/dual_profile_decoder_tp_builder.py
families/qwen/graph_blocks.py
families/qwen/graph_ops.py
families/qwen/model.py
families/qwen/native_kv_attention_builder.py
families/qwen/native_kv_contract.py
families/qwen/parallel.py
families/qwen/quantization.py
families/qwen/requirements.txt
families/qwen/runtime/CMakeLists.txt
families/qwen/runtime/bpe_tokenizer.cpp
families/qwen/runtime/chat_templates.cpp
families/qwen/runtime/chat_templates.h
families/qwen/runtime/distributed_runtime.cpp
families/qwen/runtime/distributed_runtime.h
families/qwen/runtime/inference_state.h
families/qwen/runtime/kv_cache.cpp
families/qwen/runtime/kv_cache.h
families/qwen/runtime/pipeline.cpp
families/qwen/runtime/pipeline.h
families/qwen/runtime/plugin.cpp
families/qwen/runtime/plugin_helpers.cpp
families/qwen/runtime/plugin_helpers.h
families/qwen/runtime/sampler.cpp
families/qwen/runtime/sampler.h
families/qwen/runtime/tensor_names.h
families/qwen/runtime/tokenizer.h
families/qwen/standard_decoder_builder.py
families/qwen/support.py
families/qwen/tests/__init__.py
families/qwen/tests/cpp/logits_trace.cpp
families/qwen/tests/cpp/native_kv_cache_contract_test.h
families/qwen/tests/cpp/test_qwen_native_kv_cache.cpp
families/qwen/tests/cpp/test_qwen_sampler.cpp
families/qwen/tests/cpp/test_qwen_tensor_names.cpp
families/qwen/tests/manifests/qwen3-0.6b-fp16-tp4.json
families/qwen/tests/manifests/qwen3-0.6b-fp16.json
families/qwen/tests/manifests/qwen3-0.6b-fp8-tp4.json
families/qwen/tests/manifests/qwen3-0.6b-fp8.json
families/qwen/tests/manifests/qwen3-0.6b-native-l0.json
families/qwen/tests/manifests/qwen3-0.6b-regression-native-kv-chunked-prefill.json
families/qwen/tests/manifests/qwen3-0.6b-topp-tp4.json
families/qwen/tests/manifests/qwen3-0.6b-topp.json
families/qwen/tests/manifests/qwen3-4b-instruct-2507-tp4.json
families/qwen/tests/manifests/qwen3-4b-instruct-2507.json
families/qwen/tests/package_smoke.json
families/qwen/tests/perf_validation.json
families/qwen/tests/runtime_receipt.py
families/qwen/tests/test_e2e.py
families/qwen/tests/test_runtime_receipt.py
families/qwen/tests/thresholds/qwen3-0.6b-bf16-native-kv-two-chunk-parity.json
families/qwen/tests/thresholds/qwen3-0.6b-fp16-native-kv-two-chunk-parity.json
families/qwen/tests/thresholds/qwen3-0.6b-fp16-tp4.json
families/qwen/tests/thresholds/qwen3-0.6b-fp16.json
families/qwen/tests/thresholds/qwen3-0.6b-fp8-tp4.json
families/qwen/tests/thresholds/qwen3-0.6b-fp8.json
families/qwen/tests/thresholds/qwen3-0.6b-native-l0.json
families/qwen/tests/thresholds/qwen3-0.6b-regression-native-kv-chunked-prefill.json
families/qwen/tests/thresholds/qwen3-0.6b-topp-tp4.json
families/qwen/tests/thresholds/qwen3-0.6b-topp.json
families/qwen/tests/thresholds/qwen3-4b-instruct-2507-tp4.json
families/qwen/tests/thresholds/qwen3-4b-instruct-2507.json
families/qwen/utils.py
families/qwen3_5/__init__.py
families/qwen3_5/checkpoint_mapper.py
families/qwen3_5/config.py
families/qwen3_5/graph_blocks.py
families/qwen3_5/graph_ops.py
families/qwen3_5/model.py
families/qwen3_5/runtime/CMakeLists.txt
families/qwen3_5/runtime/bpe_tokenizer.cpp
families/qwen3_5/runtime/chat_templates.cpp
families/qwen3_5/runtime/chat_templates.h
families/qwen3_5/runtime/hybrid_state.cpp
families/qwen3_5/runtime/hybrid_state.h
families/qwen3_5/runtime/inference_state.h
families/qwen3_5/runtime/kv_cache.cpp
families/qwen3_5/runtime/kv_cache.h
families/qwen3_5/runtime/pipeline.cpp
families/qwen3_5/runtime/pipeline.h
families/qwen3_5/runtime/plugin.cpp
families/qwen3_5/runtime/plugin_helpers.cpp
families/qwen3_5/runtime/plugin_helpers.h
families/qwen3_5/runtime/qwen3_5_recurrent_step_contracts.h
families/qwen3_5/runtime/recurrent_output_initializers.h
families/qwen3_5/runtime/recurrent_state.cpp
families/qwen3_5/runtime/recurrent_state.h
families/qwen3_5/runtime/sampler.cpp
families/qwen3_5/runtime/sampler.h
families/qwen3_5/runtime/tokenizer.h
families/qwen3_5/support.py
families/qwen3_5/tests/__init__.py
families/qwen3_5/tests/cpp/test_qwen3_5_recurrent_output_initializers.cpp
families/qwen3_5/tests/manifests/qwen35-9b.json
families/qwen3_5/tests/test_e2e.py
families/qwen3_5/tests/thresholds/qwen35-9b.json
families/qwen3_8/__init__.py
families/qwen3_8/checkpoint_mapper.py
families/qwen3_8/config.py
families/qwen3_8/engine_builder.py
families/qwen3_8/graph_blocks.py
families/qwen3_8/graph_ops.py
families/qwen3_8/model.py
families/qwen3_8/requirements.txt
families/qwen3_8/runtime/CMakeLists.txt
families/qwen3_8/runtime/bpe_tokenizer.cpp
families/qwen3_8/runtime/chat_templates.cpp
families/qwen3_8/runtime/chat_templates.h
families/qwen3_8/runtime/hybrid_state.cpp
families/qwen3_8/runtime/hybrid_state.h
families/qwen3_8/runtime/inference_state.h
families/qwen3_8/runtime/kv_cache.cpp
families/qwen3_8/runtime/kv_cache.h
families/qwen3_8/runtime/pipeline.cpp
families/qwen3_8/runtime/pipeline.h
families/qwen3_8/runtime/plugin.cpp
families/qwen3_8/runtime/plugin_helpers.cpp
families/qwen3_8/runtime/plugin_helpers.h
families/qwen3_8/runtime/qwen3_8_recurrent_step_contracts.h
families/qwen3_8/runtime/recurrent_output_initializers.h
families/qwen3_8/runtime/recurrent_state.cpp
families/qwen3_8/runtime/recurrent_state.h
families/qwen3_8/runtime/sampler.cpp
families/qwen3_8/runtime/sampler.h
families/qwen3_8/runtime/tokenizer.h
families/qwen3_8/support.py
families/qwen3_8/tests/__init__.py
families/qwen3_8/tests/cpp/test_qwen3_8_recurrent_output_initializers.cpp
families/qwen3_8/tests/cpp/test_qwen3_8_recurrent_pipeline.cpp
families/qwen3_8/tests/manifests/qwen38-27b.json
families/qwen3_8/tests/test_checkpoint_mapper.py
families/qwen3_8/tests/test_e2e.py
families/qwen3_8/tests/test_engine_builder.py
families/qwen3_8/tests/thresholds/qwen38-27b.json
families/qwen3_omni/__init__.py
families/qwen3_omni/checkpoint_mapper.py
families/qwen3_omni/code2wav_builder.py
families/qwen3_omni/config.py
families/qwen3_omni/graph_blocks.py
families/qwen3_omni/graph_ops.py
families/qwen3_omni/model.py
families/qwen3_omni/requirements.txt
families/qwen3_omni/runtime/CMakeLists.txt
families/qwen3_omni/runtime/audio_plan.h
families/qwen3_omni/runtime/bpe_tokenizer.cpp
families/qwen3_omni/runtime/kv_cache.cpp
families/qwen3_omni/runtime/kv_cache.h
families/qwen3_omni/runtime/pipeline.cpp
families/qwen3_omni/runtime/pipeline.h
families/qwen3_omni/runtime/plugin.cpp
families/qwen3_omni/runtime/plugin_helpers.cpp
families/qwen3_omni/runtime/plugin_helpers.h
families/qwen3_omni/runtime/residual_multinomial_kernel.cu
families/qwen3_omni/runtime/residual_multinomial_kernel.h
families/qwen3_omni/runtime/sampler.cpp
families/qwen3_omni/runtime/sampler.h
families/qwen3_omni/runtime/tokenizer.h
families/qwen3_omni/support.py
families/qwen3_omni/talker_builder.py
families/qwen3_omni/talker_weight_mapper.py
families/qwen3_omni/tests/__init__.py
families/qwen3_omni/tests/cpp/test_qwen3_omni_audio_plan.cpp
families/qwen3_omni/tests/cpp/test_qwen3_omni_sampler.cpp
families/qwen3_omni/tests/manifests/qwen3-omni-30b-a3b-instruct.json
families/qwen3_omni/tests/test_bf16_constants.py
families/qwen3_omni/tests/test_e2e.py
families/qwen3_omni/tests/test_reference_contract.py
families/qwen3_omni/tests/thresholds/qwen3-omni-30b-a3b-instruct.json
families/qwen_image/__init__.py
families/qwen_image/checkpoint_mapper.py
families/qwen_image/config.py
families/qwen_image/graph_blocks.py
families/qwen_image/graph_ops.py
families/qwen_image/model.py
families/qwen_image/qwen25_vl_text_encoder_builder.py
families/qwen_image/qwen_image_bundle_config.py
families/qwen_image/qwen_image_dit_builder.py
families/qwen_image/qwen_image_preprocessor.py
families/qwen_image/qwen_image_vae_builder.py
families/qwen_image/qwen_vl_vision_builder.py
families/qwen_image/requirements.txt
families/qwen_image/runtime/CMakeLists.txt
families/qwen_image/runtime/bpe_tokenizer.cpp
families/qwen_image/runtime/pipeline.cpp
families/qwen_image/runtime/pipeline.h
families/qwen_image/runtime/plugin.cpp
families/qwen_image/runtime/plugin_helpers.cpp
families/qwen_image/runtime/plugin_helpers.h
families/qwen_image/runtime/preprocessor_weights_helpers.h
families/qwen_image/runtime/qwen_image_batch_utils.h
families/qwen_image/runtime/qwen_image_scheduler.cpp
families/qwen_image/runtime/qwen_image_scheduler.h
families/qwen_image/runtime/qwen_image_types.cpp
families/qwen_image/runtime/qwen_image_types.h
families/qwen_image/runtime/tokenizer.h
families/qwen_image/support.py
families/qwen_image/tests/__init__.py
families/qwen_image/tests/cpp/qwen_image_e2e_harness.cpp
families/qwen_image/tests/manifests/qwen-image-2512.json
families/qwen_image/tests/manifests/qwen-image-edit-2511.json
families/qwen_image/tests/manifests/qwen-image-l0.json
families/qwen_image/tests/manifests/qwen-image.json
families/qwen_image/tests/test_e2e.py
families/qwen_image/tests/test_parity_contract.py
families/qwen_image/tests/thresholds/qwen-image-2512.json
families/qwen_image/tests/thresholds/qwen-image-edit-2511.json
families/qwen_image/tests/thresholds/qwen-image-l0.json
families/qwen_image/tests/thresholds/qwen-image.json
families/qwen_image/utils.py
families/qwen_moe/__init__.py
families/qwen_moe/checkpoint_mapper.py
families/qwen_moe/config.py
families/qwen_moe/default_decoder.py
families/qwen_moe/default_dual_profile_decoder.py
families/qwen_moe/graph_blocks.py
families/qwen_moe/graph_ops.py
families/qwen_moe/model.py
families/qwen_moe/parallel.py
families/qwen_moe/runtime/CMakeLists.txt
families/qwen_moe/runtime/bpe_tokenizer.cpp
families/qwen_moe/runtime/chat_templates.cpp
families/qwen_moe/runtime/chat_templates.h
families/qwen_moe/runtime/distributed_runtime.cpp
families/qwen_moe/runtime/distributed_runtime.h
families/qwen_moe/runtime/inference_state.h
families/qwen_moe/runtime/kv_cache.cpp
families/qwen_moe/runtime/kv_cache.h
families/qwen_moe/runtime/pipeline.cpp
families/qwen_moe/runtime/pipeline.h
families/qwen_moe/runtime/plugin.cpp
families/qwen_moe/runtime/plugin_helpers.cpp
families/qwen_moe/runtime/plugin_helpers.h
families/qwen_moe/runtime/sampler.cpp
families/qwen_moe/runtime/sampler.h
families/qwen_moe/runtime/tensor_names.h
families/qwen_moe/runtime/tokenizer.h
families/qwen_moe/support.py
families/qwen_moe/tests/__init__.py
families/qwen_moe/tests/cpp/logits_trace.cpp
families/qwen_moe/tests/manifests/qwen3-moe-30b-a3b-tp4.json
families/qwen_moe/tests/manifests/qwen3-moe-30b-a3b.json
families/qwen_moe/tests/manifests/qwen3-moe-tiny-random.json
families/qwen_moe/tests/test_e2e.py
families/qwen_moe/tests/thresholds/qwen3-moe-30b-a3b-tp4.json
families/qwen_moe/tests/thresholds/qwen3-moe-30b-a3b.json
families/qwen_moe/tests/thresholds/qwen3-moe-tiny-random.json
families/qwen_moe/tp_builder.py
families/qwen_moe/utils.py
families/qwen_vl/__init__.py
families/qwen_vl/checkpoint_mapper.py
families/qwen_vl/config.py
families/qwen_vl/decoder_tp_builder.py
families/qwen_vl/default_decoder.py
families/qwen_vl/default_dual_profile_decoder.py
families/qwen_vl/graph_blocks.py
families/qwen_vl/graph_ops.py
families/qwen_vl/lora.py
families/qwen_vl/model.py
families/qwen_vl/parallel.py
families/qwen_vl/qwen_vl_vision_builder.py
families/qwen_vl/requirements.txt
families/qwen_vl/runtime/CMakeLists.txt
families/qwen_vl/runtime/bpe_tokenizer.cpp
families/qwen_vl/runtime/cuda_stream.h
families/qwen_vl/runtime/decoded_image.h
families/qwen_vl/runtime/image_preprocessor.cpp
families/qwen_vl/runtime/image_preprocessor.h
families/qwen_vl/runtime/image_transform_helper.h
families/qwen_vl/runtime/inference_state.h
families/qwen_vl/runtime/kv_cache.cpp
families/qwen_vl/runtime/kv_cache.h
families/qwen_vl/runtime/lora_peft_artifact.cpp
families/qwen_vl/runtime/lora_peft_artifact.h
families/qwen_vl/runtime/lora_peft_loader.cpp
families/qwen_vl/runtime/lora_peft_loader.h
families/qwen_vl/runtime/lora_runtime.cpp
families/qwen_vl/runtime/lora_runtime.h
families/qwen_vl/runtime/pipeline.cpp
families/qwen_vl/runtime/pipeline.h
families/qwen_vl/runtime/plugin.cpp
families/qwen_vl/runtime/plugin_helpers.cpp
families/qwen_vl/runtime/plugin_helpers.h
families/qwen_vl/runtime/sampler.cpp
families/qwen_vl/runtime/sampler.h
families/qwen_vl/runtime/tensor_names.h
families/qwen_vl/runtime/tokenizer.h
families/qwen_vl/support.py
families/qwen_vl/tests/__init__.py
families/qwen_vl/tests/manifests/qwen25vl-3b-tp2.json
families/qwen_vl/tests/manifests/qwen25vl-3b.json
families/qwen_vl/tests/manifests/qwen3-vl-2b-tp4.json
families/qwen_vl/tests/manifests/qwen3-vl-2b.json
families/qwen_vl/tests/test_e2e.py
families/qwen_vl/tests/test_split_decoder_contract.py
families/qwen_vl/tests/test_vision_oracle.py
families/qwen_vl/tests/thresholds/qwen25vl-3b-tp2.json
families/qwen_vl/tests/thresholds/qwen25vl-3b.json
families/qwen_vl/tests/thresholds/qwen3-vl-2b-tp4.json
families/qwen_vl/tests/thresholds/qwen3-vl-2b.json
families/qwen_vl/tests/vision_oracle.py
families/qwen_vl/utils.py
families/roberta/__init__.py
families/roberta/checkpoint_mapper.py
families/roberta/config.py
families/roberta/encoder_builder.py
families/roberta/graph_ops.py
families/roberta/model.py
families/roberta/parallel.py
families/roberta/runtime/CMakeLists.txt
families/roberta/runtime/bpe_tokenizer.cpp
families/roberta/runtime/pipeline.cpp
families/roberta/runtime/pipeline.h
families/roberta/runtime/plugin.cpp
families/roberta/runtime/plugin_helpers.cpp
families/roberta/runtime/plugin_helpers.h
families/roberta/runtime/tokenizer.h
families/roberta/runtime/unigram_tokenizer.cpp
families/roberta/support.py
families/roberta/tests/__init__.py
families/roberta/tests/manifests/camembert-base-tp4.json
families/roberta/tests/manifests/camembert-base.json
families/roberta/tests/manifests/roberta-base-tp4.json
families/roberta/tests/manifests/roberta-base.json
families/roberta/tests/manifests/roberta-large-tp4.json
families/roberta/tests/manifests/roberta-large.json
families/roberta/tests/manifests/xlm-roberta-base-tp4.json
families/roberta/tests/manifests/xlm-roberta-base.json
families/roberta/tests/test_e2e.py
families/roberta/tests/test_tokenizer_contract.py
families/roberta/tests/thresholds/camembert-base-tp4.json
families/roberta/tests/thresholds/camembert-base.json
families/roberta/tests/thresholds/roberta-base-tp4.json
families/roberta/tests/thresholds/roberta-base.json
families/roberta/tests/thresholds/roberta-large-tp4.json
families/roberta/tests/thresholds/roberta-large.json
families/roberta/tests/thresholds/xlm-roberta-base-tp4.json
families/roberta/tests/thresholds/xlm-roberta-base.json
families/roberta/tp_builder.py
families/rwkv/__init__.py
families/rwkv/checkpoint_mapper.py
families/rwkv/config.py
families/rwkv/graph_ops.py
families/rwkv/model.py
families/rwkv/parallel.py
families/rwkv/runtime/CMakeLists.txt
families/rwkv/runtime/bpe_tokenizer.cpp
families/rwkv/runtime/chat_templates.cpp
families/rwkv/runtime/chat_templates.h
families/rwkv/runtime/distributed_runtime.cpp
families/rwkv/runtime/distributed_runtime.h
families/rwkv/runtime/inference_state.h
families/rwkv/runtime/pipeline.cpp
families/rwkv/runtime/pipeline.h
families/rwkv/runtime/plugin.cpp
families/rwkv/runtime/plugin_helpers.cpp
families/rwkv/runtime/plugin_helpers.h
families/rwkv/runtime/recurrent_output_initializers.h
families/rwkv/runtime/recurrent_state.cpp
families/rwkv/runtime/recurrent_state.h
families/rwkv/runtime/rwkv_recurrent_step_contracts.h
families/rwkv/runtime/sampler.cpp
families/rwkv/runtime/sampler.h
families/rwkv/runtime/tokenizer.h
families/rwkv/support.py
families/rwkv/tests/__init__.py
families/rwkv/tests/cpp/test_rwkv_recurrent_output_initializers.cpp
families/rwkv/tests/manifests/rwkv-169m-tp4.json
families/rwkv/tests/manifests/rwkv-169m.json
families/rwkv/tests/perf_validation.json
families/rwkv/tests/test_e2e.py
families/rwkv/tests/thresholds/rwkv-169m-tp4.json
families/rwkv/tests/thresholds/rwkv-169m.json
families/rwkv/tp_builder.py
families/sam/__init__.py
families/sam/checkpoint_mapper.py
families/sam/config.py
families/sam/graph_ops.py
families/sam/model.py
families/sam/parallel.py
families/sam/requirements.txt
families/sam/runtime/CMakeLists.txt
families/sam/runtime/decoded_image.h
families/sam/runtime/plugin.cpp
families/sam/runtime/plugin_helpers.cpp
families/sam/runtime/plugin_helpers.h
families/sam/runtime/sam_image_preprocess_seam.h
families/sam/runtime/sam_output_selection.h
families/sam/runtime/sam_pipeline.cpp
families/sam/runtime/sam_pipeline.h
families/sam/runtime/sam_postprocess_seam.h
families/sam/runtime/sam_prompt_seam.h
families/sam/runtime/sam_types.h
families/sam/sam_tp_builder.py
families/sam/support.py
families/sam/tests/__init__.py
families/sam/tests/manifests/sam-vit-base-tp4.json
families/sam/tests/manifests/sam-vit-base.json
families/sam/tests/test_e2e.py
families/sam/tests/test_reference_contract.py
families/sam/tests/thresholds/sam-vit-base-tp4.json
families/sam/tests/thresholds/sam-vit-base.json
families/sam2/__init__.py
families/sam2/checkpoint_mapper.py
families/sam2/float_math.py
families/sam2/graph_ops.py
families/sam2/image_builder.py
families/sam2/model.py
families/sam2/model_config.py
families/sam2/requirements.txt
families/sam2/runtime/CMakeLists.txt
families/sam2/runtime/plugin.cpp
families/sam2/runtime/sam2_bbox_postprocess.cpp
families/sam2/runtime/sam2_bbox_postprocess.h
families/sam2/runtime/sam2_device_workspace.cpp
families/sam2/runtime/sam2_device_workspace.h
families/sam2/runtime/sam2_engine_contract.h
families/sam2/runtime/sam2_mask_postprocess_cuda.cu
families/sam2/runtime/sam2_mask_postprocess_cuda.h
families/sam2/runtime/sam2_native_video_processor.cpp
families/sam2/runtime/sam2_native_video_processor.h
families/sam2/runtime/sam2_pipeline.cpp
families/sam2/runtime/sam2_pipeline.h
families/sam2/runtime/sam2_preprocess.cpp
families/sam2/runtime/sam2_preprocess.h
families/sam2/runtime/sam2_preprocess_cuda.cu
families/sam2/runtime/sam2_preprocess_cuda.h
families/sam2/support.py
families/sam2/tests/__init__.py
families/sam2/tests/cpp/operational_probe.cpp
families/sam2/tests/l4_local_runner.py
families/sam2/tests/local/README.md
families/sam2/tests/local/sam2-l4-local.json
families/sam2/tests/local/sam2-l4-local.thresholds.json
families/sam2/tests/manifests/sam2-public-core-l0.json
families/sam2/tests/operational_oracle.py
families/sam2/tests/test_e2e.py
families/sam2/tests/test_l4_local_runner.py
families/sam2/tests/test_operational_oracle.py
families/sam2/tests/thresholds/sam2-public-core-l0.json
families/sam2/tracker_builder.py
families/sam3/__init__.py
families/sam3/checkpoint_mapper.py
families/sam3/config.py
families/sam3/core_builder.py
families/sam3/graph_ops.py
families/sam3/model.py
families/sam3/requirements.txt
families/sam3/runtime/CMakeLists.txt
families/sam3/runtime/bpe_tokenizer.cpp
families/sam3/runtime/plugin.cpp
families/sam3/runtime/plugin_helpers.cpp
families/sam3/runtime/plugin_helpers.h
families/sam3/runtime/sam3_config.h
families/sam3/runtime/sam3_pipeline.cpp
families/sam3/runtime/sam3_pipeline.h
families/sam3/runtime/sam3_video_kernels.cu
families/sam3/runtime/sam3_video_kernels.h
families/sam3/runtime/sam3_video_processor.cpp
families/sam3/runtime/sam3_video_processor.h
families/sam3/runtime/sam3_video_session.cpp
families/sam3/runtime/sam3_video_session.h
families/sam3/runtime/tokenizer.h
families/sam3/support.py
families/sam3/tests/__init__.py
families/sam3/tests/manifests/sam3.json
families/sam3/tests/test_e2e.py
families/sam3/tests/test_oracle_contract.py
families/sam3/tests/test_runtime_contract.py
families/sam3/tests/thresholds/sam3.json
families/sam3/text_encoder_builder.py
families/sam3/timing_cache.py
families/sam3/tokenizer_contract.py
families/sam3/tracker_attention_builder.py
families/sam3/tracker_builder.py
families/sam3/tracker_decoder_builder.py
families/sam3/tracker_memory_builder.py
families/sam3/tracker_weights.py
families/sam3/vision_encoder_builder.py
families/sana_wm/__init__.py
families/sana_wm/builder_lifetime.py
families/sana_wm/checkpoint_mapper.py
families/sana_wm/components/__init__.py
families/sana_wm/components/gemma/__init__.py
families/sana_wm/components/gemma/checkpoint_mapper.py
families/sana_wm/components/gemma/config.py
families/sana_wm/components/gemma/dual_profile_decoder_builder.py
families/sana_wm/components/gemma/graph_blocks.py
families/sana_wm/components/gemma/graph_ops.py
families/sana_wm/components/gemma/plugin.py
families/sana_wm/components/gemma/standard_decoder_builder.py
families/sana_wm/components/ltx_video/__init__.py
families/sana_wm/components/ltx_video/checkpoint_mapper.py
families/sana_wm/components/ltx_video/graph_ops.py
families/sana_wm/components/ltx_video/ltx_dit_builder.py
families/sana_wm/components/ltx_video/ltx_vae_builder.py
families/sana_wm/config.py
families/sana_wm/model.py
families/sana_wm/model_config.py
families/sana_wm/native_plugin_builder.py
families/sana_wm/native_plugins/CMakeLists.txt
families/sana_wm/native_plugins/sana_wm_gdn_creator.cpp
families/sana_wm/native_plugins/sana_wm_gdn_plugin.cu
families/sana_wm/native_plugins/sana_wm_gdn_plugin.h
families/sana_wm/native_plugins/sana_wm_torch_conv2d_plugin.cpp
families/sana_wm/native_plugins/torch_cuda_bfloat16_math.cpp
families/sana_wm/native_plugins/torch_cuda_bfloat16_math.h
families/sana_wm/refiner_dit_builder.py
families/sana_wm/refiner_text_connector_builder.py
families/sana_wm/requirements.txt
families/sana_wm/runtime/CMakeLists.txt
families/sana_wm/runtime/pipeline.cpp
families/sana_wm/runtime/pipeline.h
families/sana_wm/runtime/plugin.cpp
families/sana_wm/runtime/plugin_helpers.cpp
families/sana_wm/runtime/plugin_helpers.h
families/sana_wm/runtime/sana_wm_bpe_tokenizer.cpp
families/sana_wm/runtime/sana_wm_scheduler_helpers.h
families/sana_wm/runtime/tokenizer.h
families/sana_wm/stage1_dit_builder.py
families/sana_wm/support.py
families/sana_wm/tests/__init__.py
families/sana_wm/tests/assets/demo_0.txt
families/sana_wm/tests/assets/demo_0_intrinsics.npy
families/sana_wm/tests/manifests/sana-wm-bidirectional.json
families/sana_wm/tests/test_e2e.py
families/sana_wm/tests/thresholds/sana-wm-bidirectional.json
families/segformer/__init__.py
families/segformer/checkpoint_mapper.py
families/segformer/config.py
families/segformer/graph_ops.py
families/segformer/model.py
families/segformer/parallel.py
families/segformer/requirements.txt
families/segformer/runtime/CMakeLists.txt
families/segformer/runtime/decoded_image.h
families/segformer/runtime/plugin.cpp
families/segformer/runtime/plugin_helpers.cpp
families/segformer/runtime/plugin_helpers.h
families/segformer/runtime/segformer_postprocess_cuda.cu
families/segformer/runtime/segformer_postprocess_cuda.h
families/segformer/runtime/segformer_postprocess_seam.h
families/segformer/runtime/segformer_preprocess_seam.cpp

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

🔍 Trigger review

Move each model into a self-owned build, runtime, dependency, and test slice behind the shared bundle and abstract Task contracts. Keep BYOK, benchmarks, examples, DevToolkit lifecycle, K2-Horizon, and current MoGe behavior as one-way consumers or family-owned implementations. Remove legacy registries, compatibility paths, and custom content digests. BREAKING CHANGE: existing bundles and integrations must be rebuilt against the new public contracts. Signed-off-by: yifeif-nv <277870278+yifeif-nv@users.noreply.github.com>

yifeif-nv added the run-internal-ci Maintainer-approved dispatch to internal CI label Aug 29, 2026

github-actions Bot removed the run-internal-ci Maintainer-approved dispatch to internal CI label Aug 29, 2026

yifeif-nv force-pushed the agent/ai-native-horizontal-scale branch from 5a8e111 to 8c77059 Compare August 29, 2026 23:29

yifeif-nv added the run-internal-ci Maintainer-approved dispatch to internal CI label Aug 29, 2026

github-actions Bot removed the run-internal-ci Maintainer-approved dispatch to internal CI label Aug 29, 2026

yifeif-nv force-pushed the agent/ai-native-horizontal-scale branch from 8c77059 to 8253660 Compare August 29, 2026 23:49

yifeif-nv added the run-internal-ci Maintainer-approved dispatch to internal CI label Aug 30, 2026

github-actions Bot removed the run-internal-ci Maintainer-approved dispatch to internal CI label Aug 30, 2026

yifeif-nv force-pushed the agent/ai-native-horizontal-scale branch from 8253660 to e0be104 Compare August 30, 2026 00:09

yifeif-nv added the run-internal-ci Maintainer-approved dispatch to internal CI label Aug 30, 2026

github-actions Bot removed the run-internal-ci Maintainer-approved dispatch to internal CI label Aug 30, 2026

yifeif-nv force-pushed the agent/ai-native-horizontal-scale branch from e0be104 to 3daa7c0 Compare August 30, 2026 00:36

yifeif-nv added the run-internal-ci Maintainer-approved dispatch to internal CI label Aug 30, 2026

github-actions Bot removed the run-internal-ci Maintainer-approved dispatch to internal CI label Aug 30, 2026

yifeif-nv force-pushed the agent/ai-native-horizontal-scale branch from 3daa7c0 to 3d07318 Compare August 30, 2026 00:56

yifeif-nv added the run-internal-ci Maintainer-approved dispatch to internal CI label Aug 30, 2026

github-actions Bot removed the run-internal-ci Maintainer-approved dispatch to internal CI label Aug 30, 2026

yifeif-nv force-pushed the agent/ai-native-horizontal-scale branch from 3d07318 to 474cbe1 Compare August 30, 2026 01:44

yifeif-nv added the run-internal-ci Maintainer-approved dispatch to internal CI label Aug 30, 2026

github-actions Bot removed the run-internal-ci Maintainer-approved dispatch to internal CI label Aug 30, 2026

yifeif-nv force-pushed the agent/ai-native-horizontal-scale branch from 474cbe1 to 89bf793 Compare August 30, 2026 07:04

yifeif-nv added the run-internal-ci Maintainer-approved dispatch to internal CI label Aug 30, 2026

yifeif-nv force-pushed the agent/ai-native-horizontal-scale branch from 10241b6 to 202a7f9 Compare August 30, 2026 09:38

yifeif-nv force-pushed the agent/ai-native-horizontal-scale branch from 505f7c6 to de21139 Compare August 31, 2026 17:57

yifeif-nv force-pushed the agent/ai-native-horizontal-scale branch from ce3bd64 to d7fbc38 Compare September 1, 2026 20:21

yifeif-nv added the run-internal-ci Maintainer-approved dispatch to internal CI label Sep 1, 2026

github-actions Bot removed the run-internal-ci Maintainer-approved dispatch to internal CI label Sep 1, 2026

yifeif-nv force-pushed the agent/ai-native-horizontal-scale branch from f591438 to 0b63f0c Compare September 2, 2026 23:26

This was referenced Sep 3, 2026

ci(families): add isolation review guardrails yifeif-nv/TensorRT-Model-Connect-fork#2

Closed

ci(families): add isolation review guardrails #1139

Merged

yifeif-nv force-pushed the agent/ai-native-horizontal-scale branch 4 times, most recently from 7016cea to 6415529 Compare September 4, 2026 21:15

yifeif-nv force-pushed the agent/ai-native-horizontal-scale branch from 6415529 to 9777fef Compare September 5, 2026 08:27

Uh oh!

Conversation

yifeif-nv commented Aug 29, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Background

Result

Preserved Functionality

Deliberate Simplifications

Coverage Preservation

Validation

Remaining Gate

Review Notes

Change categories

Risk level

Uh oh!

coderabbitai Bot commented Aug 29, 2026

Uh oh!

coderabbitai Bot commented Aug 29, 2026

Uh oh!

coderabbitai Bot commented Aug 29, 2026

Uh oh!

coderabbitai Bot commented Aug 30, 2026

Uh oh!

coderabbitai Bot commented Aug 30, 2026

Uh oh!

coderabbitai Bot commented Aug 30, 2026

Uh oh!

coderabbitai Bot commented Aug 30, 2026

Uh oh!

coderabbitai Bot commented Aug 30, 2026

Uh oh!

coderabbitai Bot commented Aug 30, 2026

Uh oh!

coderabbitai Bot commented Aug 30, 2026

Uh oh!

coderabbitai Bot commented Aug 31, 2026

Uh oh!

coderabbitai Bot commented Aug 31, 2026

Uh oh!

coderabbitai Bot commented Sep 1, 2026

Uh oh!

coderabbitai Bot commented Sep 1, 2026

Uh oh!

coderabbitai Bot commented Sep 1, 2026

Uh oh!

github-actions Bot commented Sep 1, 2026

Uh oh!

coderabbitai Bot commented Sep 1, 2026

Uh oh!

coderabbitai Bot commented Sep 2, 2026

Uh oh!

coderabbitai Bot commented Sep 2, 2026

Uh oh!

coderabbitai Bot commented Sep 2, 2026

Uh oh!

coderabbitai Bot commented Sep 2, 2026

Uh oh!

coderabbitai Bot commented Sep 2, 2026

Uh oh!

coderabbitai Bot commented Sep 2, 2026

Uh oh!

coderabbitai Bot commented Sep 2, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Review skipped

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

yifeif-nv commented Aug 29, 2026 •

edited

Loading

coderabbitai Bot commented Sep 2, 2026 •

edited

Loading