3 people 903d2efe7e [Attention][MLA] Fuse Kimi-K3 chunked-context K/V packing (#51772) ...

Signed-off-by: Yongye Zhu <zyy1102000@gmail.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

2026-08-12 22:42:57 -07:00

..

deepseek_v32

[Perf] Optimize DeepSeek V3.2 sequence parallelism (#51434)

2026-08-07 16:51:01 -07:00

fixtures

[BUGFIX] Fix Pixtral consolidated format vision weight loading (#39916)

2026-04-19 22:25:03 -07:00

inkling

[2/N][Feat][Perf] Add new warmup infrastructure for JITs. Add predicate filtering for JIT warmup, and migrate Inkling FA4 (#49315)

2026-08-11 07:46:07 -04:00

kimi_k3

[Attention][MLA] Fuse Kimi-K3 chunked-context K/V packing (#51772)

2026-08-12 22:42:57 -07:00

language

[ROCm] Remove stale SDPA and skinny GEMM workarounds (#50907)

2026-08-11 14:19:16 -04:00

multimodal

[CI] Mirror external test assets in vLLM S3 (#52064)

2026-08-13 02:08:59 +00:00

quantization

[Migration] Migrate bitsandbytes support to OOT plugin (#43529)

2026-08-09 10:27:31 +08:00

transformers

Bump Transformers version to 5.15.0 (#51668)

2026-08-11 09:00:51 -07:00

__init__.py

[CI/Build] Move test_utils.py to tests/utils.py (#4425)

2024-05-13 23:50:09 +09:00

registry.py

[Model] Add native Dots3 NOTE multimodal support (#51255)

2026-08-12 23:13:20 +08:00

test_adapters.py

[BugFix][Pooling] Skip weight-prefix probe when model has WeightsMapper (#50890)

2026-08-05 21:12:33 -07:00

test_deepseek_v4_mega_moe.py

[Model] Support R3 capture with DeepGEMM MegaMoE (#51831)

2026-08-12 01:07:07 -07:00

test_dspark_mla.py

[K3 Perf] Optimize k3 dspark fused kv, 4.5~4.6x kernel performance improvement (#50585)

2026-08-07 12:16:11 -07:00

test_initialization.py

[ROCm] Run init test engine in-process to avoid KV-cache OOM (#48527)

2026-07-15 06:39:31 +00:00

test_qwen3_5_mtp_config.py

[Bugfix][Model] Fix Qwen3.5 MTP for text-only checkpoints (#50734)

2026-08-10 11:04:53 +00:00

test_registry.py

[Model] Add native Dots3 NOTE multimodal support (#51255)

2026-08-12 23:13:20 +08:00

test_terratorch.py

[CI/Build] Skip Prithvi/Terratorch model-registry tests when terratorch is missing (#41389)

2026-04-30 12:47:55 -07:00

test_utils.py

[Test] Refactor hard coded device string in test files under compile/quantization/models/model_executor folders (#38901)

2026-04-15 11:02:35 +08:00

test_vision.py

replace batch_norm to numerically identical without cudnn (#51734)

2026-08-11 12:43:20 +08:00

utils.py

[Core] Keep attention backends eligible for text-only serving of prefix-LM models (#48796)

2026-07-25 03:06:14 -07:00