Skip to content

v0.78.0-dev20260819

Pre-release
Pre-release

Choose a tag to compare

View all tags
@github-actions github-actions released this 19 Aug 09:03
· 465 commits to main since this release
Immutable release. Only release title and notes can be modified.
v0.78.0-dev20260819
a85c79a
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

Note

If you are installing from a release, please refer to the README, INSTALLATION instructions, and any other documentation packaged with the release, not on the main branch. There may be differences between the latest main and the previous release.

The changelog will now follow, showing the changes from last release.

This release was generated by the CI workflow https://github.com/tenstorrent/tt-metal/actions/runs/32200610835

LLK (low-level kernels)

  • LLK e2e: skip SyncFull dest handoff for tiny-tile matmul occupancy > 8 PR 53502
  • Add tests for blaze-authored SFPU activation/filter LLKs PR 53123
  • Limit LLK tilize polluter negative controls to same-format cases PR 52226
  • Document FP16B unity constant PR 53503
  • trigger quasar build on fuser changes PR 52069
  • ci: update tt-exalens version to 0.3.31 in LLK test harness PR 53514
  • Promote the hw_cleanup LLK family to experimental PR 53296
  • Implement cumsum for Quasar PR 52935
  • BH LLK topk_xl: unfused SFPLOADMACRO primitives + fused-u16 end-to-end family, with LLK-level tests PR 53455
  • BH LLK topk: replay-recorded step loads/stores for sort phases >= 4 (1.15-1.2x single-core topk) PR 53454
  • Quasar: Allocate buffer descriptors during operation initialization PR 52762
  • Promote custom_mm and compressed_custom_mm LLK families to experimental PR 52727

Metalium (tt-metal core)

  • Preserve descriptor-backed simulator topology classification PR 52785
  • Reset the shared GO mailboxes from the last CQ, once the others have drained PR 52777
  • Set KERNEL_COMPILE_TIME_ARG_MAP via a generated header instead of -D define PR 53404
  • Added wait command to unused mesh device cq to eliminate race condition PR 52782
  • Set resolved_dispatch_core_type in the offline compile enumerator PR 53185
  • Test Only: run sharded DRAM readback on any available bank count PR 53192
  • Per-core H2D socket FIFO in hybrid mode PR 53322
  • Automated UMD Bump 18.08.2026 PR 53479
  • Promote the hw_cleanup LLK family to experimental PR 53296
  • BH LLK topk_xl: unfused SFPLOADMACRO primitives + fused-u16 end-to-end family, with LLK-level tests PR 53455
  • Replace the GMG_UNGROUPED_TOP8 define with a defaultless template parameter PR 53075
  • Feature: add single stage support for pipeline builder PR 52866
  • Have DFB get_read/write_ptr() APIs return the uncached address ranges on Quasar DM PR 52769
  • Quasar: Cache hw_thread_idx in a thread_local variable PR 53384
  • Quasar: Allocate buffer descriptors during operation initialization PR 52762
  • Rewire demo tree onto canonical deepseek_compute_kernel_hw_startup.h PR 53294
  • Promote custom_mm and compressed_custom_mm LLK families to experimental PR 52727

TT-NN

  • ttnn.topk multi-core: fix silent value corruption for >32 flattened rows PR 53453
  • #52680: Fix noc_async_{read,write}_sharded per-chunk size underflow PR 52154
  • #52681: Clamp tilize retile shrink-case output tile-rows to buffer capacity PR 52180
  • Add accurate fp32 SFPU path for ttnn.max via fast_and_approximate_mode PR 52718
  • #51235: Fix reduce_to_root intermediate tensor spec and c_18 circular buffer overflow PR 52852
  • Fix permute precision loss PR 53410
  • Add fp32 sort index test PR 53327
  • prune the reduce test grid and re-enable test_reduction.py on ttsim PR 53360
  • test(graph_query): assert softmax DFB peak instead of CB peak PR 53446
  • MultiDeviceGlobalSemaphore removal PR 53213
  • group_norm: synthesize per-group masks in the kernel PR 48640
  • Remove TT_METAL_QSR_TILIZE_UNPACK_TO_DEST PR 53339
  • #9147: Add optional output tensor support for ttnn::repeat PR 52686
  • Replace LLK workaround with (now working) mul_reuse_dest_tiles PR 53419
  • Relax TTNN op numeric tolerances PR 45240
  • Add sigmoid-gated RMSNorm PR 52783
  • high_bw_all_gather: restore bandwidth for batch-indexed gathers PR 53368
  • Replace the GMG_UNGROUPED_TOP8 define with a defaultless template parameter PR 53075
  • Feature: add single stage support for pipeline builder PR 52866
  • Have DFB get_read/write_ptr() APIs return the uncached address ranges on Quasar DM PR 52769
  • Quasar: Allocate buffer descriptors during operation initialization PR 52762
  • Rewire demo tree onto canonical deepseek_compute_kernel_hw_startup.h PR 53294

Models

  • Restrict Blackhole DeepSeek HCA end-to-end tests PR 53511
  • Set the LayerAck env in Prefill Runner test PR 53495
  • DeepSeek V4 HCA: functional prefill PR 52008
  • group_norm: synthesize per-group masks in the kernel PR 48640
  • prefill: drop the runner self-test / KV validation PR 52214
  • Remove TT_METAL_QSR_TILIZE_UNPACK_TO_DEST PR 53339
  • Migrate BGE-M3 to the tiered Models CI as tier 2, and retire (Single-card) Demo tests PR 53042
  • Replace the GMG_UNGROUPED_TOP8 define with a defaultless template parameter PR 53075
  • Kimi K2.6 Dflash in prefill runner PR 52802
  • Rewire demo tree onto canonical deepseek_compute_kernel_hw_startup.h PR 53294
  • Promote custom_mm and compressed_custom_mm LLK families to experimental PR 52727

Infrastructure & CI

  • Restrict Blackhole DeepSeek HCA end-to-end tests PR 53511
  • Added wait command to unused mesh device cq to eliminate race condition PR 52782
  • Set resolved_dispatch_core_type in the offline compile enumerator PR 53185
  • prune the reduce test grid and re-enable test_reduction.py on ttsim PR 53360
  • ci: fix compile-time-tracker push auth broken by actions/checkout v6 PR 53385
  • DeepSeek V4 HCA: functional prefill PR 52008
  • Split codeowners-group-analysis: extract owner-selection into a Python script PR 52410
  • trigger quasar build on fuser changes PR 52069
  • Migrate BGE-M3 to the tiered Models CI as tier 2, and retire (Single-card) Demo tests PR 53042
  • prefill: move sc4 KV-PCC leg budget to shield team PR 53491
  • Update owner_id in models_e2e_tests.yaml PR 53536
  • emule: add wormhole+blackhole smoke-test script PR 49636
  • Add sigmoid-gated RMSNorm PR 52783
  • add jonathansuTT to /models/ codeowners PR 53449
  • Move deepseek blitz (sc4, sc16) and wan (sc4) tests to pacakge and release PR 53509
  • Expand ttsim SKUs to mirror per-topology HW SKUs PR 53100
  • Kimi K2.6 Dflash in prefill runner PR 52802
  • Alphabetize CODEOWNERS entries (ownership-preserving) PR 53560

Documentation

  • docs: remove stale ttnn demos page PR 53450

Tooling

  • Update tt-exalens version to 0.3.30 PR 53505

Other

  • sweeps: replay stale traced topk vectors without their indices tensor PR 53161
  • Allow overriding x86 -march for Ivy Bridge hosts PR 49041
  • Migrate BGE-M3 to the tiered Models CI as tier 2, and retire (Single-card) Demo tests PR 53042
Assets 27
Loading