vllm-project/vllm

Pull Requests Commits

ywang96 committed 32 days ago

6a7b85d9

Merge branch 'main' into fix-qwen3-parse

ywang96 committed 33 days ago

Verified 9f7ee892

[CI][AMD][BugFix] Skip tests in test_unquantized_backend_selection that should not run on ROCm (#34655)

rasmith committed 33 days ago

Verified fcd6ac97

[Model Runner V2] Minor simplification for DCP (#34786)

WoosukKwon committed 33 days ago

Verified 95be2a7f

[Bugfix] Remove assert causing hipErrorStreamCaptureUnsupported (#34455)

JadenMathias committed 33 days ago

Verified 0e60c925

[Misc] Add mooncake-transfer-engine to kv_connectors requirements (#34826)

stmatengss committed 33 days ago

Verified d7ff2220

Merge branch 'main' into fix-qwen3-parse

ywang96 committed 33 days ago

Verified 47262501

[Bugfix] Redo Qwen3.5/Qwen3-Next GDN projector fusion (#34697)

Isotr0py committed 33 days ago

Verified c0bd8b13

[Bugfix] Fix NVFP4 TRTLLM MoE non-gated support; add gsm8k for Nemotron-3-Nano FP8+NVFP4 (#34725)

mgoin committed 33 days ago

Verified caeb887b

[CI][Bugfix] Fix multinode test script (#34820)

ilmarkov committed 33 days ago

Verified 6b3166a7

[CI] temporarily disable multi-node tests (#34825)

robertgshaw2-redhat committed 33 days ago

Verified 25e2e136

[Model Bash] DeepSeek R1 BF16 Min Latency QKV A GEMM (0.5% E2E Speedup) (#34758)

robertgshaw2-redhat committed 33 days ago

Verified 6874638b

Add unit tests for fp8 output fusion of triton_attn (#34228)

bringlein committed 33 days ago

Verified e24663c5

[Model Runner V2] Avoid prepare prefill kernel launch overhead (#34780)

njhill committed 33 days ago

Verified c50e105a

[Renderer] Deprecate code paths for old input processing (#34775)

DarkLight1337 committed 33 days ago

Verified a766b303

[Quantization] - Added uses_meta_device_weights to quant config (#34645)

Josephasafg committed 33 days ago

Verified 1faa8cb7

[Bugfix] fix activation in cpu_fused_moe_torch call (#34696)

michalowski-arm committed 33 days ago

Verified e89a91d9

[Bugfix] Fix prefix creation for Qwen3.5 (#34723)

mgoin committed 33 days ago

Verified 909b1471

[Bugfix] Fix quant RMS norm fusion for quantization with TMA-aligned scales (#33255)

ElizaWszola committed 33 days ago

Verified a88b3be7

Merge branch 'main' into fix-qwen3-parse

ywang96 committed 33 days ago

Verified b58a393f

ywang96 committed 33 days ago

ec5bf35c

[Model Runner V2] A bit more PP simplification (#34766)

njhill committed 33 days ago

Verified a49ea5a5

Merge branch 'main' into fix-qwen3-parse

ywang96 committed 33 days ago

Verified 522f51d7

ywang96 committed 33 days ago

47b93195

add more models

ywang96 committed 33 days ago

31b4eff4

ywang96 committed 33 days ago

12204cf8

[CI/Build] Remove use of `skip_v1` (#34699)

DarkLight1337 committed 33 days ago

Verified 30ebe0dc

[ROCm][CI] Removed hard-coded attn backend requirement for Qwen VL (#34753)

AndreasKaratzas committed 33 days ago

Verified cef65f07

[Core] Fix SSRF bypass via backslash-@ URL parsing inconsistency (#34743)

russellb committed 33 days ago

Verified 6f3b2047

[torch.compile] Turn on silu+fp4 quant fusion by default for O1+ (#34718)

ProExpertProg committed 33 days ago

Verified 02e8f26c

Older