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 264216 次贡献
更少更多
- frostbyte_neo 推送了仓库 github-featured/vllm-project--vllm 的 skip-lmfe-tests 分支bf316c2650 [CI] Skip lm-format-enforcer test cases
- frostbyte_neo 推送了仓库 github-featured/vllm-project--vllm 的 simon-mo-patch-1 分支eca0895879 [Build] Update build image to ubuntu 22
- frostbyte_neo 推送了仓库 github-featured/vllm-project--vllm 的 running-deque 分支90eb28ca21 [V1][Scheduler] Use dict for running queue
- frostbyte_neo 推送了仓库 github-featured/vllm-project--vllm 的 rocm_silu_mul_quant 分支比较 3 提交 »cbd94b205f fix benchmark007219cffd move stuff aroundfd97715675 Use fused silu-mul-quant with BatchedTritonExperts
- frostbyte_neo 推送了仓库 github-featured/vllm-project--vllm 的 revert-batch-kv-cache-swap-38460 分支3da0bdec77 Revert "[Perf] Batch KV cache swap copies via cuMemcpyBatchAsync (#38460)"
- frostbyte_neo 推送了仓库 github-featured/vllm-project--vllm 的 revert-50434-fix-supported-quant 分支ebc2cdf4aa Revert "[XPU] [BugFix] Add deepseek_v4_fp8 to xpu supported_quantization list…"
- frostbyte_neo 推送了仓库 github-featured/vllm-project--vllm 的 revert-40687-matthias.skinny-gemm-n5 分支8ad3817a8c Revert "[ROCm][Perf] Support N=5 in wvSplitK skinny GEMM kernels for speculat…"
- frostbyte_neo 推送了仓库 github-featured/vllm-project--vllm 的 revert-27600-torch-utils-import 分支f68f7ee030 Revert "[nit]: Fix import for the lmcache integration (#27600)"
- frostbyte_neo 推送了仓库 github-featured/vllm-project--vllm 的 revert-26740-wentao-optimize-startup-log-2 分支63bd5018a1 Revert "[Log] Optimize Startup Log (#26740)"
- frostbyte_neo 推送了仓库 github-featured/vllm-project--vllm 的 revert-22299-main 分支42018e8d96 Revert "Implicit language-model-only mode via limit-mm-per-prompt (#22299)"