[DOC] Add fuse_minimax_qk_norm (#39782)

Jee Jee Li committed 7 months ago

[Refactor] Drop direct dependency on librosa (#39079)

Nick Cao committed 7 months ago

[ZenCPU] AMD Zen CPU Backend with supported dtypes via zentorch weekly (#39967)

Chinmay-Kulkarni-AMD committed 7 months ago

[Bugfix] Fix k_proj's bias for GLM-ASR (#40160)

Rishapveer Singh committed 7 months ago

[Doc] Fix outdated source reference comment in anthropic/serving.py (#40189)

z1ying committed 7 months ago

[Frontend] Add multimodal support to /inference/v1/generate endpoint (#38405)

Nithin Chalapathi committed 7 months ago

[Doc] Add Realtime Transcription section to supported_models.md (#39845)

z1ying committed 7 months ago

[Core] Reduce mm scheduler, get_num_embed overhead (#40143)

milesial committed 7 months ago

[XPU] fix all_reduce all-zero accuracy issue under torch.compile (#39844)

Chaojun Zhang committed 7 months ago

[CI] Speed up test_fused_marlin_moe (#40178)

Michael Goin committed 7 months ago

[XPU]fake impl for xpu fp8_gemm (#39984)

Xinyu Chen committed 7 months ago

Remove outdated tests test_mixtral_moe and test_duplicated_ignored_sequence_group (#40175)

Michael Goin committed 7 months ago

[Kernel] Add MXFP4 W4A4 CUTLASS MoE kernel for SM100 (#37463)

Michael Goin committed 7 months ago

[Kernel] [Helion] Force disable HOP path due to performance regression (#40171)

Yanan Cao committed 7 months ago

[ROCm] Fix TurboQuant on ROCm: backend routing, flash-attn compat, int64 overflow (#39953)

aditi-amd committed 7 months ago

[AMD][CI] Update DeepEP branch (#38396)

Ryan Rock committed 7 months ago

[ROCm][CI] Build fastsafetensors from source so it links against libamdhip64 (#39978)

Andreas Karatzas committed 7 months ago

skip fp8e4b15 on xpu (#39957)

Xinyu Chen committed 7 months ago

[Misc][UX] Map mimo reasoning and tooling parsers (#40089)

Roger Wang committed 7 months ago

feat: Add LoRA support for Gemma4ForConditionalGeneration (#39291)

allgather committed 7 months ago