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 12, 12:15 UTC
main
[refactor] CTConfig methods to static/class methods (#28870)
HDCharles committed 12 months ago
[Perf] Optimize batch invariant BMM, 18.1% Throughput improvement, 10.7% TTFT improvement (#29345)
Wentao Ye committed 12 months ago
Revert "[Bugfix] Fix GPT-OSS AR+NORM fusion (#28841)" (#29483)
Huamin Li committed 12 months ago
Optimize the wording of the document and unify the terminology and th… (#29491)
yxt committed 12 months ago
Fix TeleChatForCausalLM not register issue (#29473)
Yejing Lai committed 12 months ago
[Bugfix] Fix handling of image embeds in models (#29480)
Cyrus Leung committed 12 months ago
[Misc] Allow LM only loading for Pixtral (#29451)
Roger Wang committed 12 months ago
[Performance][MLA][ROCm] Remove redundant D2D copy in deepseek (#27457)
Pleaplusone committed 12 months ago
[Kernel] Use pre-allocated output buffer for triton kernel fused_experts (#29219)
Xin Yang committed 12 months ago
Bump actions/checkout from 4 to 6 (#29293)
dependabot[bot] committed 12 months ago
[UX] Suppress gloo log spam (#29250)
Michael Goin committed 12 months ago
[Misc] Add backup hash algorithm for FIPS constrained environments (#28795)
George D. Torres committed 12 months ago
dummy run corner case (#29433)
Xieyang Xu committed 12 months ago
fix assertion for single world use case (uni) (#29429)
Lucia Fang committed 12 months ago
[BugFix] Fix initialization of draft model. (#29319)
Andrey Khalyavin committed 12 months ago
Change warning logs to debug for unimplemented MXFP4 Linear/Attention (#29441)
Michael Goin committed 12 months ago
[caching] Add enable_prompt_embeds and cpu_offload_gb to compile hashes. (#29435)
Zhengxu Chen committed 12 months ago
[Core] Support logprobs with spec decode + async scheduling (#29223)
Nick Hill committed 12 months ago
[Compile] Refactor. Move PostGradPassManager out of Compilation config (#29340)
Ilya Markov committed 12 months ago
Attempt to fix GPU OOM in a spec-decoding test (#29419)
Eldar Kurtić committed 12 months ago