12月1月2月3月4月5月6月7月8月9月10月11月
周一周三周五
一年内 264202 次贡献
更少更多
- frostbyte_neo 推送了仓库 github-featured/vllm-project--vllm 的 v0.8.0 分支比较 9 提交 »966f933ee1 [Bugfix] Fix LoRA extra vocab size (#15047)1a504aff6c [Bugfix] Fix broken CPU quantization due to triton import (#15038)01ca85bbd8 [MODEL] Add support for Zamba2 models (#13185)d82b9487ea [Bugfix] Register serializers for V0 MQ Engine (#15009)be13281d4b [Bugfix] Loosen type check to avoid errors in V1 (#15021)
- frostbyte_neo 推送了仓库 github-featured/vllm-project--vllm 的 v0.16.0-torch291 分支比较 8 提交 »a976961fb7 [ROCm][CI] Pin TorchCodec to v0.10.0 for ROCm compatibility (#34447)772f708391 [ci] Use the right tag for CPU arm64 image (#34915)9030895929 [CI/Build] Fix gRPC version mismatch (#35013)61c344652b img build31575c1da0 Revert "img build"
- frostbyte_neo 推送了仓库 github-featured/vllm-project--vllm 的 v0.16.0-cu128 分支比较 3 提交 »e4e900feeb merge commit 3a3851a7487 remove wheel build dependency on extensionsc55c7f67fd merge commit 1
- frostbyte_neo 推送了仓库 github-featured/vllm-project--vllm 的 upgrade-transformers-compressed-tensors 分支bcb5e98f11 [Dependencies] Upgrade transformers to >= 5.5.0 and compressed-tensors to >= 0.15.0
- frostbyte_neo 推送了仓库 github-featured/vllm-project--vllm 的 update_from_kv_xfer_finished_race_fix 分支比较 2 提交 »2797adb329 cleanup7477823407 [Bugfix] Fix race condition when KV transfer times out before request finishes
- frostbyte_neo 推送了仓库 github-featured/vllm-project--vllm 的 ucx_oneshot_ar 分支b0ed553028 [Distributed] Add UCX one-shot AllReduce for DP metadata sync
- frostbyte_neo 推送了仓库 github-featured/vllm-project--vllm 的 tpu_v1_optimized 分支比较 7 提交 »70b4e46e70 compilation is fixed5fb9dbe6f6 fix capture model996b92ccb4 swap works!2b0526fa15 works!7be649256f fixes