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 2, 19:30 UTC
main
[Model] Use AutoWeightsLoader for InternLM2 (#38278)
Javier De Jesus committed 5 months ago
[Reasoning] [Bugfix] Reject invalid thinking_token_budget values (#43402)
linzm1007 committed 5 months ago
[Misc] Support interleaved custom image benchmark datasets (#43636)
Thibault Castells committed 5 months ago
[Misc][Refactor][ROCm] Convert MoRI-related envvars to extra config args (#43303)
Simon Danielsson committed 5 months ago
[KV Transfer] Enable HMA by default for connectors that support it (#41847)
Ethan Feng committed 5 months ago
[Bugfix] Apply fc_norm in Eagle3DeepseekV2 combine_hidden_states (#43482)
Yubo Wang committed 5 months ago
[chores][log] change registry log from `warning` to `debug` (#43045)
Hank_ committed 5 months ago
Add CuTe DSL sparse compressor support (#43584)
Jie Fang committed 5 months ago
Upgrade tpu-inference to v0.20.0 (#43394)
Sting Lin committed 5 months ago
[CPU] Enable non-divisible GQA for decode workitems in mixed batches (#43032)
zhao, zhenhui committed 5 months ago
[GDN] GDN Prefill kernel for SM100 (#43273)
Thien Tran committed 5 months ago
[Bugfix][Model] Fix GPT2ForSequenceClassification sub-module prefix (#43579)
Huanyu Yang committed 5 months ago
[Frontend] Split the offline inference APIs and utils. (#43553)
wang.yuqi committed 5 months ago
[Bugfix] fix device mismatch in MiniCPM-o-4_5 resampler (#43194)
Yan Ma committed 5 months ago
[KV Connector] Propagate MooncakeStore load failures (#42788)
Dao007forever committed 5 months ago
[KV Connector][Bugfix] MooncakeStore: don't double-apply Eagle prune in load_mask (#43516)
Dao007forever committed 5 months ago
[DeepSeek V4] Move MegaMoE input prep kernel to nvidia/ops (#43632)
Woosuk Kwon committed 5 months ago
[LoRA] Add one shot triton kernel For MoE LoRA (#42290)
Jee Jee Li committed 5 months ago
[XPU] Ensure RNG offset alignment with PyTorch requirements in XPU sampler (#43028)
Chaojun Zhang committed 5 months ago
[Kernel] Remove NormGateLinear (#43554)
Jee Jee Li committed 5 months ago