[Cleanup] Refactor profiling env vars into a CLI config (#29912)

Benjamin Chislett committed 11 months ago

[Model Runner V2] Support num NaNs in logits (#30187)

Woosuk Kwon committed 11 months ago

[Model Runner V2] Fix Triton warning on tl.where (#30355)

Woosuk Kwon committed 11 months ago

[BugFix] Fix non detected failing tests (#30277)

Ilya Markov committed 11 months ago

Update AMD test definitions (2025-12-08) (#30298)

Alexei-V-Ivanov-AMD committed 11 months ago

[Compile] Fix torch warning `TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled` (#29897)

Wentao Ye committed 11 months ago

[BugFix] Fix `assert batch_descriptor.num_tokens == num_tokens_padded` (#30173)

Lucas Wilkinson committed 11 months ago

[Feature] Batch-Invariant Support for FA2 and LoRA (#30018)

quanliu committed 11 months ago

[BUGFIX] Mistral tool call parser v11+ (#30332)

Julien Denize committed 11 months ago

[ROCm] Aiter Quant Kernels (#25552)

vllmellm committed 11 months ago

[Bugfix]: Fix glm46 awq marlin moe wna16 compatibility (#30210)

Dongjie Zou committed 11 months ago

[bugfix][quantization] Fix fp8 per_tensor scale shape (#30257)

haoyangli-amd committed 11 months ago

[Structured Output][Reasoning] Improves decoding throughput for models using single-token reasoning endings. (#30056)

Hubert de La Jonquiere committed 11 months ago

[DCP][Bugfix][CI] Fix accuracy issue of DCP when using FLASH_ATTN_MLA (#30309)

Jaya Yuan committed 11 months ago

[Bugfix] Qwen 3 VL Embedding loading (#30303)

wang.yuqi committed 11 months ago

[CI] Fix Flaky test_eagle_max_len Test (#30306)

Micah Williamson committed 11 months ago

[Attention] Make `split_decodes_and_prefills(..., require_uniform=True)` support padding (#29644)

Lucas Wilkinson committed 11 months ago

[Misc] Fix safetensors import for safe_open (#30300)

Yongtao Huang committed 11 months ago

[Model][Quantization] Restore MoE + GGUF models support (incl. Qwen3 MoE) by allowing Sideload Parameters (#30116)

Tsukasa OI committed 11 months ago

[Doc] update Intel GPU MM status in Feature x Hardware matrix (#30294)

Fanli Lin committed 11 months ago