1. 11 Mar, 2026 2 commits
  2. 10 Mar, 2026 1 commit
  3. 04 Mar, 2026 2 commits
  4. 03 Mar, 2026 1 commit
  5. 02 Mar, 2026 1 commit
  6. 26 Feb, 2026 1 commit
  7. 24 Feb, 2026 1 commit
    • laibao's avatar
      • perf(v1): 增加可选的快速 token-id 拷贝路径 · d3a95d54
      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
      d3a95d54
  8. 08 Feb, 2026 1 commit
  9. 06 Feb, 2026 1 commit
  10. 05 Feb, 2026 1 commit
  11. 02 Feb, 2026 1 commit
  12. 27 Jan, 2026 1 commit
  13. 26 Jan, 2026 3 commits
  14. 25 Jan, 2026 1 commit
  15. 24 Jan, 2026 4 commits
  16. 23 Jan, 2026 4 commits
  17. 22 Jan, 2026 3 commits
  18. 21 Jan, 2026 3 commits
  19. 20 Jan, 2026 7 commits
  20. 19 Jan, 2026 1 commit