Skip to content

0.8.0

Choose a tag to compare

View all tags
@github-actions github-actions released this 20 Oct 22:44
0.8.0
c0fd758
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

What's Changed

🚨 Breaking Changes

  • refactor(evaluator): move Harbor runtime modules into runtimes/harbor by @ngoncharenko in #2444
  • refactor(evaluator)!: HarborRunnerTarget selects its agent from a source union by @SandyChapman in #2455
  • fix(evaluator): propagate Gym env secrets through a shared source by @ngoncharenko in #2469
  • refactor!: delete the generated nemo_helix SDK by @maxdubrinsky in #2635
  • refactor(evaluator)!: GymRunnerTarget selects its agent from a source by @SandyChapman in #2482
  • feat(cli)!: replace nemo agent with nemo describe by @mikeknep in #2738
  • feat(optimization)!: add strategy titles and requires_config; rename legacy to config-tuning by @stefan-kickoff in #2910

✨ New Features

  • feat(evaluator): add Harbor taskset integration test and notebook by @ngoncharenko in #2136
  • feat(studio): Add QuickstartPanel for sandbox Storybook by @nakolean in #2403
  • feat(evaluator-sdk): run a complete Fabric config in the Harbor agent by @SandyChapman in #2357
  • feat(optimization): pass judge inference params to tunable_rag_evaluator by @marcusds in #2330
  • feat(agents): expose enabled agent deployment methods [ASTD-668] by @marcusds in #2385
  • feat(evaluator): propagate env_secrets from the Harbor runner by @ngoncharenko in #2399
  • feat: orchestrate between the numeric and prompt optimizations by @gabwow in #1860
  • feat(example-plugin): show the shared CLI pattern in middleware-configs commands by @mikeknep in #2457
  • feat(customizer): serve saved job templates from the customization router by @htolentino-nvidia in #2099
  • feat(studio): Customizer/Data designer start page redesign by @htolentino-nvidia in #2238
  • feat(studio): Banner for Sample workspace by @nakolean in #2456
  • feat(agent-optimization): show strategy descriptions in list-strategies by @mikeknep in #2492
  • feat(studio): rename agent creation to registration [ASTD-681] by @marcusds in #2521
  • feat(studio): turn on FF VITE_FF_AGENT_OVERVIEW_ENABLED by @nv-odrulea in #2524
  • feat(studio): save a customization job as a template, and start from one by @htolentino-nvidia in #2516
  • feat(evaluator): evaluate a registered agent through the Fabric runner by @SandyChapman in #2320
  • feat(models): add family and model_providers filters to list models by @marcusds in #2537
  • feat(evaluator): evaluate a registered agent through the Harbor runner by @SandyChapman in #2358
  • feat(studio): Sample quickstart dashboard by @nakolean in #2514
  • feat(jobs): add OpenShell-compatible nhx-tasks image variant by @maxdubrinsky in #2496
  • feat(agents): add openshell deployment mode on OpenShell 0.1.2 by @maxdubrinsky in #2473
  • feat(scaled-evals): Files-backed artifact storage backend + tests (1/4) by @benmccown in #2081
  • feat(evaluator): live metric evaluation endpoint by @nv-odrulea in #2400
  • feat(studio): trim the agent Details tab [ASTD-682] by @marcusds in #2522
  • feat(scaled-evals): route artifact storage through the Files service (2/4) by @benmccown in #2122
  • feat(studio): deploy agents on create with a method the platform can run [ASTD-659] by @marcusds in #2425
  • feat(evaluator): propagate Fabric env secrets in localhost, sandbox, job runs by @ngoncharenko in #2510
  • feat(plugin): add deployments HTTP consumer client by @benmccown in #2172
  • feat(skills): add release test scope workflow by @vevaidyanath-lang in #2173
  • feat(builder): core entities and contracts by @albcui in #2363
  • feat(builder): compile a build request into a job and submit by @albcui in #2364
  • feat(studio): Optimization Form behind FF by @steramae-nvidia in #2051
  • feat(intake): represent failed evaluator results by @SandyChapman in #2593
  • feat(studio): pick the eval wizard dataset from a fileset [ASTD-721] by @marcusds in #2651
  • feat(studio): Deploy HPO job parameters in new agent by @steramae-nvidia in #2649
  • feat(builder): signing and completion by @albcui in #2431
  • feat(studio): build an agent image before deploying to Docker or Kubernetes [ASTD-714] by @marcusds in #2616
  • feat(builder): build plane steps with sandbox supervisor by @albcui in #2432
  • feat(builder): run it end to end on minikube by @albcui in #2509
  • feat(evaluator): evaluate a registered agent through the Gym runner by @SandyChapman in #2490
  • feat(setup): add api endpiont to create sample workspace and demo agent by @mmogallapalli in #2663
  • feat(switchyard): add the switchyard optimization strategy by @stefan-kickoff in #2591
  • feat(nhx-sandbox): add minimal shared sandbox package by @JashG in #2474
  • feat(agents): add a prompt-master optimization plugin by @stefan-kickoff in #2229
  • feat(evaluator): FieldMapping allows array element matches, with limts by @nv-odrulea in #2577
  • feat(cli): add searchable setup model picker by @mckornfield in #2622
  • feat(studio): show a blocking screen when NeMo Helix is unreachable by @aahunt-nv in #2623
  • feat(studio): accept Parquet and CSV datasets in Run evaluation [ASTD-749] by @marcusds in #2667
  • feat(scaled-evals): install Harbor runners from per-version requirements with OpenSandbox support by @JashG in #2475
  • feat(rename): add reusable plugin rename profiles and Evals invocation by @gabwow in #2665
  • feat(studio): Enable submission of optimization job through UI by @steramae-nvidia in #2656
  • feat(agents): add remote agent fabric adapter by @mmogallapalli in #2687
  • feat(studio): Support options for optimization strategies by @steramae-nvidia in #2680
  • feat(studio): deploy HPO trials that tune an agent's system instructions by @marcusds in #2700
  • feat(nemo-agents): add Pi (nvidia.fabric.pi) harness support by @benmccown in #2653
  • feat(studio): send agent ethos with analysis runs and add Run analysis now by @marcusds in #2718
  • feat(evaluator): filter tasks and tasksets by kind, metadata, metrics, members, tags and text by @SandyChapman in #2513
  • feat(agents): add support for nooa fabric adapter by @mmogallapalli in #2698
  • feat(scaled-evals): add OpenSandbox Harbor environment and ownership-scoped cleanup by @JashG in #2476
  • feat(agent-hardener): install by default and rename the menu entry to Agent Hardening [ASTD-755] by @marcusds in #2696
  • feat(builder): prefix job names, return image_ref, and filter images by set by @albcui in #2721
  • feat(studio): Add sample workspace creation modal by @nakolean in #2724
  • feat(scaled-evals): add harbor_opensandbox runtime backend by @JashG in #2477
  • feat(cli): Remove --agent-mode by @mikeknep in #2740
  • feat(anonymizer): run GLiNER PII detector in-process via Files HF pull-through by @benmccown in #2709
  • feat(studio): link trace agent names to agent page by @nv-odrulea in #2759
  • feat(studio): Artifact/logs in opti job details by @steramae-nvidia in #2761
  • feat(cli): CLI agents help command cleanup by @stefan-kickoff in #2757
  • feat(auth): broker confidential OIDC login for CLI by @ironcommit in #2545
  • feat(studio): cherry-pick: draft a customization job with AI (#2654) by @htolentino-nvidia in #2794
  • feat(studio): use primary/secondary CTA buttons in dashboard quicksta… by @steramae-nvidia in #2834
  • feat(insights): add evidence notes to AnalystResult by @mikeknep in #2874
  • feat(studio): cherry-pick: compare evaluation runs side by side on the experiment page (#2817) by @marcusds in #2932
  • feat(studio): cherry-pick: explain why a finished customization job is still active (#2838) by @walston in #2912
  • feat(studio): cherry-pick: enable trace graph feature flag by default (#2792) [ASTD-767] by @marcusds in #2943
  • feat(jobs): OpenShell job backend for CPU tasks by @maxdubrinsky in #2638
  • feat(studio): cherry-pick: plot tokens on the experiment Pareto chart (#2765) by @marcusds in #2934
  • feat(studio): cherry-pick: add an evaluation scorecard to the experiment page (#2819) by @marcusds in #2935
  • feat(nemo-agents): forward per-model extensions metadata to Fabric (backport to 0.8) by @benmccown in #3000
  • feat(studio): support server-side OIDC sessions by @ironcommit in #2733
  • feat(studio): add a job details action to fine-tuning rows by @aray12 in #2960
  • feat(studio): show in-progress fine-tuning runs above the model table by @aray12 in #2961

🔧 Improvements

  • refactor(entities): consolidate LoRA composite formatting; drop dead parser by @benmccown in #2196
  • perf(inference-gateway): skip no-op model-entity-map rebuilds by @benmccown in #2171
  • refactor(cli): plugin CLI helpers by @mikeknep in #2418
  • refactor(guardrails,intake): import CLI helpers from nemo_helix_plugin by @mikeknep in #2458
  • refactor(scaled-evals): rename api/s3.py -> api/artifacts.py (3/4) by @benmccown in #2123
  • perf(studio): lazy-load plugin nav icons instead of bundling all of lucide by @marcusds in #2607
  • refactor(models): read managed deployments via the deployments API by @benmccown in #2195
  • refactor(models): migrate create/delete to the deployments API by @benmccown in #2202
  • refactor: move services, plugins, and tests off the generated SDK to typed clients by @maxdubrinsky in #2600
  • perf(cli): defer heavy plugin-primitive imports to speed up CLI startup by @benmccown in #2643
  • refactor(tools): move license and vendor tooling to tools/nemo-helix-tools by @maxdubrinsky in #2647
  • perf(cli): reuse platform HTTP transports by @ironcommit in #2630
  • refactor(plugin-cli): remove legacy generated verbs by @mckornfield in #2776

🐛 Bug Fixes

  • fix(studio): show a tooltip for truncated Anonymizer model dropdown values [NMP-232] by @marcusds in #2374
  • fix(studio): metric scores description should not be optional by @nv-odrulea in #2381
  • fix(cli): flatten auditor audit job command by @ironcommit in #2377
  • fix(agents): repair session chat streaming and run plugin tests in CI by @mikeknep in #2412
  • fix(example-plugin): use flat generated verbs by @ironcommit in #2386
  • fix: reject invalid NGC metadata before publishing by @svvarom in #2345
  • fix(plugin): keep entity metadata when typed clients parse responses by @mikeknep in #2416
  • fix(plugin): use the global base URL and context for generated submit commands by @mikeknep in #2417
  • fix(studio): seed guardrail configs with the default model's entity name [ASTD-656] by @marcusds in #2406
  • fix(studio): default Anonymizer entity mode to auto-detect [ASTD-663] by @marcusds in #2421
  • fix(sandboxed-gym): refuse OPEN_SANDBOX_* in the sandbox env; pin NEMO_RL_REF once by @SandyChapman in #2401
  • fix(auditor): support --output-format on configs and targets commands by @mikeknep in #2459
  • fix: recover forward merge #2471 by @aire-ops[bot] in #2472
  • fix: recover forward merge #2479 by @aire-ops[bot] in #2480
  • fix(agents): use the shared output format and base URL for agents commands by @mikeknep in #2461
  • fix(models): surface provider auth failures as ERROR status by @benmccown in #2442
  • fix(insights): apply cli standardization to insights cli by @mikeknep in #2460
  • fix(auth): ignore external runtime scopes by @mckornfield in #2489
  • fix(skills): don't clobber a configured remote NHX_BASE_URL by @benmccown in #2440
  • fix(studio): flip flag to show model evaluation by @nv-odrulea in #2370
  • fix(auth): ignore external runtime scopes by @ironcommit in #2527
  • fix(rl): frame gym host errors and pin the rebased NeMo-RL by @soluwalana in #2375
  • fix: recover forward merge #2529 by @aire-ops[bot] in #2538
  • fix(scaled-evals): expose platform jobs as flat CLI by @ironcommit in #2394
  • fix(agent-hardener): use flat job CLI by @ironcommit in #2392
  • fix: recover forward merge #2535 by @aire-ops[bot] in #2540
  • fix(auth): enforce workload token exchange for agents and services by @ironcommit in #2251
  • fix(train): grpo offline fails with setuptools error by @soluwalana in #2539
  • fix: recover forward merge #2543 by @aire-ops[bot] in #2544
  • fix(cli, skills): Make skills, help, and config view more remote cluster-aware by @mikeknep in #2494
  • fix: NOOA client None backend_format by @mikeknep in #2548
  • fix(customizer): add DTensor checkpoint config to DPO by @anubhutivyas in #2551
  • fix(studio): Fix loading flicker for evaluation details by @nakolean in #2549
  • fix(train): resolve NVBUG 6484778 & 6581813 by @soluwalana in #2546
  • fix: recover forward merge #2553 by @aire-ops[bot] in #2560
  • fix(studio): Update workspaces onCreate by @nakolean in #2558
  • fix(cli): report a denied bind as not permitted, not EADDRINUSE by @maxdubrinsky in #2534
  • fix(insights): Allow service principals to read/write insights by @mikeknep in #2570
  • fix(studio): offer only the image builds and runtimes the platform supports [ASTD-684] by @marcusds in #2526
  • fix(agents): follow up on the Fabric 0.4 / Relay 0.9 upgrade by @marcusds in #2559
  • fix(evaluator): let a sandboxed Gym host survive a cold node's image pull by @SandyChapman in #2561
  • fix(evaluator): reject a whitespace-only agent-eval task id at submit by @SandyChapman in #2594
  • fix(jobs): resolve get_job by ID as well as name by @benmccown in #2488
  • fix: recover forward merge #2568 by @maxdubrinsky in #2599
  • fix(common): give platform clients the generated SDK's retry policy by @maxdubrinsky in #2598
  • fix(studio): file-preview refresh 404 + dark-mode code blocks by @maxdubrinsky in #2603
  • fix(studio): let IntakeLayout split into its own chunk [ASTD-709] by @marcusds in #2604
  • fix: stop leaking internal ticket IDs in user-facing output by @benmccown in #2573
  • fix(studio): move agent model and description into the overview Details panel by @aray12 in #2575
  • fix: recover forward merge #2619 by @aire-ops[bot] in #2621
  • fix(release): preserve version series for pinned nightlies by @svvarom in #2629
  • fix(models): mount the deployments plugin in the models and gateway integration test apps by @SandyChapman in #2639
  • fix(studio): keep recovery actions on the unauthorized workspace screen by @aray12 in #2601
  • fix: recover forward merge #2628 by @aire-ops[bot] in #2657
  • fix(studio): Show additional details while optimization job is running by @steramae-nvidia in #2655
  • fix(evaluator): stop inferring integer judge schemas from range bounds [ASTD-746] by @marcusds in #2661
  • fix(cli): warn about legacy setup directories by @mckornfield in #2618
  • fix(studio): offer Upload instead of Deploy for agents registered from traces by @walston in #2614
  • fix(studio): expand sidebar navigation sections by default by @aahunt-nv in #2690
  • fix: recover forward merge #2695 by @aire-ops[bot] in #2701
  • fix(studio): mask new value input in edit secret modal by @walston in #2684
  • fix(studio): Remove models when inference providers are deleted by @dmariali in #2658
  • fix(studio): explain the members 403 instead of retrying it by @aray12 in #2662
  • fix: make sure sidecar does not attempt to call the DB by @mckornfield in #2702
  • fix: escape percents for alembic by @mckornfield in #2705
  • fix: avoid false ATIF tool errors from returned content by @mdcox in #2717
  • fix(cli): rename agents plugins header by @stefan-kickoff in #2725
  • fix(studio): stop rendering HelixGuard temporarily by @htolentino-nvidia in #2723
  • fix(train): resolve missing head_type of retrieval models by @soluwalana in #2704
  • fix(studio): show error page instead of first use by @nakolean in #2742
  • fix(opensandbox): mount writable Harbor directories in BatchSandbox templates by @JashG in #2581
  • fix: recover forward merge #2748 by @aire-ops[bot] in #2749
  • fix(studio): Add disabled options for optimization start strategies by @steramae-nvidia in #2739
  • fix(secrets): respond with 422 on misconfigured encryption provider by @soluwalana in #2745
  • fix(evaluator): re-anchor platform-resolved loopback gateway URLs by @nv-odrulea in #2711
  • fix(studio): Update selected workspace for sample by @nakolean in #2756
  • fix(studio): reuse an existing compatible fileset when starting a template by @aray12 in #2752
  • fix: bound fileset name size, report correct training time by @soluwalana in #2753
  • fix(entities): refresh stale role bindings before denying workspace a… by @mmogallapalli in #2755
  • fix(studio): use new live endpoint to test model evals, remove temp hack by @nv-odrulea in #2758
  • fix(studio): update training loss chart empty state by @dmariali in #2746
  • fix(studio): show deployment URL for docker and k8s agent deployments by @walston in #2754
  • fix(models): keep a deployment model when its provider is deleted by @anubhutivyas in #2763
  • fix(studio): default API base URL to same-origin by @mckornfield in #2731
  • fix(agents): replace lookalike domain in email triage tutorial (0.8) by @svvarom in #2789
  • fix(agents): address port allocation race by @mmogallapalli in #2795
  • fix(agents): show container deployment logs by @mckornfield in #2808
  • fix(customization): drop submit verb from backend cli by @mckornfield in #2775
  • fix(build): serialize API dependency cache access by @svvarom in #2843
  • fix(helm): route OpenAPI and readiness endpoints by default by @mckornfield in #2847
  • fix(cli): keep pagination progress off stdout by @mckornfield in #2865
  • fix(cli): use require_workspace() in jobs watch/tail/create --wait by @benmccown in #2861
  • fix(studio): stop the model config panel corrupting builder models by @marcusds in #2867
  • fix(jobs): propagate cancellation to public child tasks by @mckornfield in #2848
  • fix(studio): default Anonymizer entity detector to the built-in in-process GLiNER2 by @benmccown in #2871
  • fix(agent-hardener): validate inputs and align release runtime wiring by @mckornfield in #2850
  • fix(studio): Fix deployment of optimization jobs by @steramae-nvidia in #2873
  • fix(studio): Fine tune config created with AI failed to be carried over to the form by @htolentino-nvidia in #2879
  • fix(cli): validate custom provider API keys during setup by @tylersbray in #2890
  • fix(entities): retry workspace deletion after cleanup failures by @tylersbray in #2889
  • fix(auth): accept legacy flat OIDC discovery documents by @ironcommit in #2891
  • fix(optimization): accept weighted sums and preserve timeout diagnostics by @mckornfield in #2849
  • fix(agents): allow task principals to read agents by @mckornfield in #2897
  • fix(cli): restore data-designer create with no flags by @mikeknep in #2913
  • fix: reject anchored optimize config paths by @mckornfield in #2922
  • fix(agents): report malformed agent YAML without a traceback by @mckornfield in #2921
  • fix(studio): Docs issues in 0.8 by @steramae-nvidia in #2948
  • fix(studio): report real tokens/sec in model chat by @marcusds in #2823
  • fix(deployments): cherry pick - correctly track startup timeout failures by @mmogallapalli in #2949
  • fix(agents): preserve subprocess startup diagnostics by @mckornfield in #2931
  • fix(docs): correct examples and add regression lint checks by @mckornfield in #2967
  • fix(optimization): attribute preflight errors to failing phase by @mckornfield in #2898
  • fix: read the release catalog from the source SHA, not main by @crookedstorm in #2991
  • fix: support vLLM for inline auto-deploy and default to model-free NIM by @anubhutivyas in #3003
  • fix(customizer): remove submit verb + unsloth docs update by @anubhutivyas in #2999
  • fix(customizer): load cross-encoder LoRA adapters without the SEQ_CLS head for merged export by @anubhutivyas in #3013
  • fix(auth): raise embedded PDP fuel and memory default caps by @benmccown in #3012
  • fix(sandboxed-gym): reap job sandboxes when a signal arrives inside a running loop by @SandyChapman in #3004
  • fix(evals): read live-eval metric secrets as evals on behalf of the caller by @SandyChapman in #3007
  • fix(jobs): generate valid auto job names and stop reporting name errors as 409 by @SandyChapman in #3006
  • fix(studio): link an adapter to the run that produced it by @aray12 in #2959
  • fix(garak): read the packaged plugin cache instead of racing on a per-user copy by @SandyChapman in #3031
  • fix(evals): keep registered-agent MCP secret templates literal through Gym's config loader by @SandyChapman in #3008
  • fix(helm): preserve legacy Postgres claim labels on 0.8 by @svvarom in #3045
  • fix(evals): resolve platform gateway URLs for Harbor containers by @gabwow in #3041
  • fix(rl): prepare writable Gym workspaces before sandbox mounts by @gabwow in #3042
  • fix(entities): keep the workspace when a resource delete fails by @tylersbray in #2951
  • fix(docs): wait for provider binding after READY in the embedding tutorial by @anubhutivyas in #3054
  • fix(agent-hardener): require agent-hardener>=0.0.16 for OpenShell 0.1.4 by @koralchapnik in #3028
  • fix(safe-synthesizer): disable dataframe check for PII replace only by @mckornfield in #3021
  • fix(entities): reclaim expired workspace deletions by @tylersbray in #2955
  • fix(docs): Trace graph docs fix by @steramae-nvidia in #3075
  • fix(studio): show evaluation scores, replies and tool calls on test pages [ASTD-828] by @marcusds in #3057
  • fix(security): update container Python and jobs-launcher Go dependencies by @mckornfield in #2947
  • fix(deployments): dial openshell sandbox services through the gateway by Host header on 0.8 by @maxdubrinsky in #3100
  • fix: properly propagate error for prompt phase by @mckornfield in #3083
  • fix(auth): make embedded PDP platform scope-family check O(1) by @anastasia-nesterenko in #3053
  • fix(agent-hardener): require agent-hardener>=0.0.17 for AgentBreaker scoring fixes by @koralchapnik in #3101
  • fix(studio): hide empty row action menus by @mckornfield in #3015
  • fix: train shm sizing, evals notebook call by @soluwalana in #3116
  • fix(local): make model and Intake endpoints reachable from Docker agents by @gabwow in #3082
  • fix(entities): report a blocked workspace delete instead of success by @tylersbray in #3119
  • fix(safe-synthesizer): line up differential privacy params with upstream by @mckornfield in #3102
  • fix(agent-hardener): publish local-config artifacts on 0.8 by @puhuim in #3086

📝 Documentation

🧰 Maintenance

  • chore: create a release notes skill by @mckornfield in #2407
  • chore: add release notes for 0.7 by @mckornfield in #2408
  • test: stop unit tests depending on free host ports, wall-clock slack, or a live harness by @SandyChapman in #2405
  • test(inference-gateway): check fixture isolation across consecutive contexts, not nested ones by @SandyChapman in #2453
  • build: run the source toolchain on Python 3.13 to match the task image by @SandyChapman in #2454
  • chore: forward merge remaining release/0.7 changes by @svvarom in #2478
  • chore: sweep docs for more nano -> lightning model refs by @mikeknep in #2515
  • ci: dispatch post-release cleanup by @svvarom in #2501
  • test(platform-runner): drive the adapters sidecar entrypoint test without a wall clock by @SandyChapman in #2520
  • chore(studio): update Kaizen UI Foundations to 1.13.0 [ASTD-620] by @marcusds in #2435
  • test: fail fast and name stuck task when TestClient teardown hangs by @mikeknep in #2523
  • chore(customizer): add magi_attention dependency in automodel by @anubhutivyas in #2327
  • chore(agents): upgrade to Fabric 0.4.0 by @mmogallapalli in #2495
  • chore(insights): follow the NeMo Compass repository rename by @callingmedic911 in #2569
  • chore(scaled-evals): drop object-store settings orphaned by the Files migration (4/4) by @benmccown in #2124
  • chore(docs): remove opencode references from skill install desc by @mmogallapalli in #2612
  • chore: add update-shell to uv tool install in docs by @mckornfield in #2609
  • chore(sdk): remove generated evaluations/experiments surface by @ironcommit in #2626
  • test(e2e): cover a wheels-v1 Gym environment that vendors only its own extras by @SandyChapman in #2641
  • chore(train): upgrade transformers in unsloth and RL by @soluwalana in #2613
  • test(nemo_helix_ext): stop the wait_for_status tests patching the global clock by @SandyChapman in #2669
  • chore: forward merge release/0.7 documentation fix by @svvarom in #2691
  • chore: forward merge release/0.7 Docker documentation by @svvarom in #2694
  • chore: forward merge release/0.7 into main by @svvarom in #2730
  • chore(agents): remove usage and leaderboard commands by @mmogallapalli in #2727
  • test(agents): add integration tests for existing harnesses by @mmogallapalli in #2706
  • chore(cli): un-hide config command and global --context option by @mikeknep in #2726
  • chore: rename NeMo Evaluator to NeMo Helix Evals by @gabwow in #2712
  • chore(auditor): rename auditor plugin to garak plugin by @parkanzky in #2692
  • test: support Platform-Deploy e2e harness by @mckornfield in #2506
  • chore(openapi): publish a single platform spec by @mckornfield in #2671
  • chore: clean up and rename garak tests by @mckornfield in #2771
  • chore(customizer): make vllm default for customizer auto-deploy + use multi-nim as default values by @anubhutivyas in #3043
  • chore: remove ffmpeg unused bits from safe-synthesizer, downgrade vLLM by @mckornfield in #3135

Other Changes

  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2397
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2398
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2411
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2420
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2422
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2436
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2441
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2443
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2470
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2471
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2479
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2507
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2518
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2528
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2530
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2531
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2532
  • Forward-merge release/0.5 into main by @rapids-bot[bot] in #2529
  • Forward-merge release/0.6 into main by @rapids-bot[bot] in #2535
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2542
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2543
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2547
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2553
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2562
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2568
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2602
  • Forward-merge release/0.6 into main by @rapids-bot[bot] in #2619
  • Forward-merge release/0.7 into main by @soluwalana in #2682
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2628
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2695
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2703
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2707
  • Forward-merge release/0.7 into main by @rapids-bot[bot] in #2748

New Contributors

Full Changelog: 0.7.3...0.8.0

Contributors

marcusds, soluwalana, and 31 other contributors
Assets 2
Loading