- 16 Mar, 2026 3 commits
- 09 Mar, 2026 1 commit
-
-
xuxz authored
-
- 06 Mar, 2026 3 commits
- 05 Mar, 2026 3 commits
- 04 Mar, 2026 3 commits
- 03 Mar, 2026 3 commits
- 02 Mar, 2026 3 commits
- 28 Feb, 2026 1 commit
-
-
yangql1 authored
-
- 26 Feb, 2026 1 commit
-
-
laibao authored
新增 VLLM_V1_USE_REDUCED_TOPK_TOPP_SAMPLER 开关并补充适用场景说明 在 V1 GPU 输入批预计算 max_top_k/has_any_no_top_k,native sampler 满足条件时走 reduced fast path,异常自动回退
-
- 25 Feb, 2026 1 commit
-
-
SAC_fanth authored
-
- 24 Feb, 2026 3 commits
-
-
SAC_fanth authored
-
laibao authored
新增 router_capture 工具,用于按 num_tokens/rank 过滤并落盘 MoE router logits 在 Qwen3MoeSparseMoeBlock 中接入采集调用,并在 torch.compile 场景下自动跳过 补充 VLLM_MOE_ROUTER_CAPTURE* 环境变量
-
laibao authored
- 新增环境变量 `VLLM_V1_FAST_TOKEN_ID_COPY`(默认关闭) - 在 `CachedRequestState` 中缓存 int32 的 prompt token ids(numpy 数组) - 开启后在 `InputBatch` 中使用 `np.copyto` 拷贝 prompt/output token ids
-
- 19 Feb, 2026 1 commit
-
-
王敏 authored
-
- 16 Feb, 2026 2 commits
- 13 Feb, 2026 1 commit
-
-
王敏 authored
-
- 12 Feb, 2026 1 commit
-
-
zhuwenwen authored
-
- 11 Feb, 2026 3 commits
- 10 Feb, 2026 4 commits
- 09 Feb, 2026 3 commits