A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 3, 16:30 UTC
main
[Bugfix] Fix EPLB initialization for VLM wrapper models (#39805)
Roy Wang committed 6 months ago
[XPU][CT] Support mxfp8 moe model (#41918)
Kunshang Ji committed 6 months ago
[Feature] Add instruction support for score/rerank chat templates (#42412)
Krish Gupta committed 6 months ago
[Bugfix][Spec Decode] Wire draft_probs into probabilistic draft_model rejection (#40269)
Siddharth Bedekar committed 6 months ago
[Bugfix] Fix Gemma4ToolParser streaming float corruption (#42128)
longguo committed 6 months ago
[Refactor] Use shared utils in hermes tool parser (#42570)
Flora Feng committed 6 months ago
fix(tool-parser): preserve "none"/"nil" strings as valid enum values in minimax_m2 (#39599)
Yiyang "Ian" Liu committed 6 months ago
[CI] set max transformers version for skywork model (#42104)
Divakar Verma committed 6 months ago
[Bugfix] V1: support tuple model outputs in ubatch wrapper (dbo + spec decode) (#40789)
Yufeng He committed 6 months ago
[Bugfix] Fix DeepSeek V4 MTP HC state handling (#42320)
Mohammad Miadh Angkad committed 6 months ago
[Core][MM] Do not use urllib3 to parse data URLs (#42535)
Lukas Geiger committed 6 months ago
expose flex block size for batch invariant mode (#41252)
liangel-02 committed 6 months ago
[Quantization] Rework quantization_config to use QuantKey and allow for activation override (#41566)
Michael Goin committed 6 months ago
[PD] Fix broken NIXL EP installation (#42542)
ovidiusm committed 6 months ago
[Frontend] add support for thinking_token_budget in completions (#42116)
Walter Beller-Morales committed 6 months ago
[CI] Fix pre-commit issue (#42563)
Wentao Ye committed 6 months ago
[ModelRunner V2] Share identical MTP weights (#42538)
Nick Hill committed 6 months ago
[Quark] Support loading Quark NVFP4 checkpoints in vLLM (#35859)
fxmarty-amd committed 6 months ago
Remove verifier model type check in speculative config (#42536)
Fynn Schmitt-Ulms committed 6 months ago
[Bugfix] Handle real-world gpt-oss tool call output in Harmony parsing (#42454)
Ben Browning committed 6 months ago