vllm-project/vllm

Pull Requests Commits

alexm-redhat committed 1 year ago

cd3ea013

[Bugfix][NIXL] Fix Async Scheduler timeout issue (#25808)

NickLucche committed 1 year ago

Verified da63274d

[Core] GC Debug callback (#24829)

Jialin committed 1 year ago

Verified c216119d

[Bug]: Set LD_LIBRARY_PATH to include the 'standard' CUDA location (#25766)

smarterclayton committed 1 year ago

Verified 5546acb4

[torch.compile]: Add VLLM_DEBUG_DUMP_PATH environment variable (#25651)

ZJY0516 committed 1 year ago

Verified c0ec8183

[Core] Refactor self.model() to call a helper for subclassing. (#25084)

patrick-toulme committed 1 year ago

Verified b65e56ba

[env] default nixl side port conflicts with kv-event zmq port (#25056)

panpan0000 committed 1 year ago

Verified 49996cd5

[docs] transcriptions API audio upload (#25446)

yyzxw committed 1 year ago

Verified ecb37e27

[Bugfix][WideEP] Apply TP Attn + EP MoE fix to other models (#24982)

tlrmchlsmth committed 1 year ago

Verified a5354b3e

[Bugfix] Fix triton import precommit failure (#25803)

tlrmchlsmth committed 1 year ago

Verified f9df8b4a

Fix GPTQ model loading in Transformers backend (#25770)

hmellor committed 1 year ago

Verified ec152c87

Add filtering for chat template kwargs (#25794)

russellb committed 1 year ago

Verified 7977e502

Validate API tokens in constant time (#25781)

russellb committed 1 year ago

Verified 3f5d902d

[Bugfix] Merge MM embeddings by index instead of token IDs (#16229)

DarkLight1337 committed 1 year ago

Verified 27d7638b

[Bugfix] Add missing `image_size` for phi4_multimodal (#25796)

Renovamen committed 1 year ago

Verified 17617398

[Misc] Update openai client example file for multimodal (#25795)

ywang96 committed 1 year ago

Verified 23b8ee67

[Misc] Fix codeowners override for v1 sample and attention (#25037)

22quinn committed 1 year ago

Verified 39391520

[CI/Build] Reorganize root-level V1 tests (#25767)

DarkLight1337 committed 1 year ago

Verified cd87bfbf

[CI/Build] Add timing to Model Executor Test (#25799)

22quinn committed 1 year ago

Verified b3613e3a

[CI/Build] Consolidate model loader tests and requirements (#25765)

DarkLight1337 committed 1 year ago

Verified d346ec69

[Bugfix] Allow Only SDPA Backend for ViT on B200 for Qwen3-VL (#25788)

yewentao256 committed 1 year ago

Verified c242c980

[Multimodal][Speculative Decoding]Eagle Eagle3 mm support, enablement on qwen2.5vl (#22872)

david6666666 committed 1 year ago

Verified f1d53d15

Add flashinfer-build.sh and register precompiled cu128 wheel in Dockerfile (#25782)

mgoin committed 1 year ago

Verified 92da847c

Add option to restrict media domains (#25783)

russellb committed 1 year ago

Verified 3958b96b

[Core] Don't count preempted tokens in prefix cache hit rate (#25787)

zhuohan123 committed 1 year ago

Verified 8bf8f458

[Spec decode] automatically disable mm for text-only draft models (#25667)

jmkuebler committed 1 year ago

Verified 6f5c0931

[Bugfix] Optimize CpuGpuBuffer initialization (#25447)

namanlalitnyu committed 1 year ago

Verified 4e33a7ea

Kernel-override Determinism [1/n] (#25603)

bwasti committed 1 year ago

Verified dc48ba0c

Reduce the Cuda Graph memory footprint when running with DBO (#25779)

SageMoore committed 1 year ago

Verified 4778b426

[spec decode] Consolidate speculative decode method name for MTP (#25232)

zixi-qi committed 1 year ago

Verified c70ac4b8

Older