cleanup at::Tag::needs_fixed_stride_order (#28974)

Boyuan Feng committed 12 months ago

[CI/Build][AMD] Fix import errors in tests/kernels/attention (#29032)

rasmith committed 12 months ago

[Bug] Fix torch dynamo warning Dynamo detected a call to a `functools.lru_cache` (#29038)

Wentao Ye committed 12 months ago

[ci][amd] fix EPLB execution test (#28742)

Bradley D committed 12 months ago

[ROCm][BugFix] Fix shared expert loading error when disable `VLLM_ROCM_USE_AITER_FUSION_SHARED_EXPERTS` (#28633)

Pleaplusone committed 12 months ago

[V0 Deprecation] Remove `num_lookahead_slots` (#29000)

Cyrus Leung committed 12 months ago

[Frontend] Allow parsed tool arguments (#28820)

Quentin Gallouédec committed 12 months ago

[Model][QwenVL] Replace `torch.repeat_interleave` with faster `np.repeat` (#28964)

Lukas Geiger committed 12 months ago

[Bugfix] Revert custom attention mask for gemma3-mm (#28995)

Isotr0py committed 12 months ago

[Bugfix] Handle broken frames in video loading (#29001)

Canlin Guo committed 12 months ago

[Bugfix] Fix precision loss in LoRA-wrapped RowParallelLinear by fusing bias into GEMM (#28972)

prashanth058 committed 12 months ago

Consolidate Nvidia ModelOpt quant config handling for all quantization methods (#28076)

Shengliang Xu committed 12 months ago

[Feat] Iteration-level profiling for Torch and CUDA profiler (#28987)

Benjamin Chislett committed 12 months ago

[cpu][ci] Add initial set of tests for Arm CPUs (#28657)

Fadi Arafeh committed 12 months ago

[AMD] Use Decoupled Kernel Block Size to Support AITER MLA block_size=1 (#27715)

Qiang Zhang committed 12 months ago

[DeepSeek + LMCache Multiprocess] handle MLA for deepseek model + LMCache Multiprocess connector (#29039)

Kuntai Du committed 12 months ago

[torchao] fix safetensors for sharding (#28169)

liangel-02 committed 12 months ago

[Misc] Colorize logs (#29017)

Nick Hill committed 12 months ago

[GC Debugger] Simply and improve GC Debugger Utils (#29029)

Jialin Ouyang committed 12 months ago

[Refactor] Optimize `select_experts` (#28069)

Wentao Ye committed 12 months ago