[Core] Get num_encoder_tokens from scheduler config (#24989)

Russell Bryant committed about 1 year ago

[misc] fix typo in value error (#24995)

Prashant Gupta committed about 1 year ago

[UX] Remove "quantization is not fully optimized yet" log (#25012)

Michael Goin committed about 1 year ago

[Docs] fix invalid doc link (#25017)

yyzxw committed about 1 year ago

[UX] Enforce valid choices for envs like VLLM_ATTENTION_BACKEND, etc (#24761)

Michael Goin committed about 1 year ago

[Benchmarks] Add MMVU video dataset support and clean up deprecated datasets (#24719)

Isotr0py committed about 1 year ago

[Core] Use `CpuGpuBuffer` for block table tensors (#24795)

Nick Hill committed about 1 year ago

[Doc] Add --force-overwrite option to generate_cmake_presets.py (#24375)

elvischenv committed about 1 year ago

Removes source compilation of nixl dependency (#24874)

Benjamin Bartels committed about 1 year ago

[FP8] Extend per-token-group quantization support to QuantFP8 (#24342)

Tahsin Tunan committed about 1 year ago

[CI] GPT-OSS GPQA eval test for Blackwell (#24920)

Michael Goin committed about 1 year ago

[CI][Bugfix] Fix failing Blackwell test (#24993)

Matthew Bonanni committed about 1 year ago

[Perf] Reuse workspace for FP8+FP4 Marlin MoE (#20500)

Michael Goin committed about 1 year ago

fp8 kv cache support fix for torch.compile (#22758)

Aleksandr Malyshev committed about 1 year ago

Use kwargs for long lists of `EngineCoreRequest` arguments in tests and fix extra kwargs (#24987)

Andrew Sansom committed about 1 year ago

[gpt-oss][1][bugfix] fix streaming final output (#24466)

Andrew Xia committed about 1 year ago

[ROCm] Add dependencies for ROCm (#24900)

Concurrensee committed about 1 year ago

[MISC] Add code owners of vllm/v1 to vllm/v1/core (#24928)

Chen Zhang committed about 1 year ago

[gpt-oss][1b] streaming add item id, content id (#24788)

Andrew Xia committed about 1 year ago

Directly get max encoder len from VLLM config in V1 (#24866)

Sugar committed about 1 year ago