github-featured / ollama--ollama

发布

0 发布 616 Git 标签

RSS 订阅源
  • v0.30.11 d26a58557d
    比较

    MLX: wire up scheduler selected context size for ps (#16918)

    frostbyte_neo 发布于 2026-06-26 15:47:03 +00:00 | 106 次提交 在此版本后已推送到 main

    In the PS output, expose the scheduler selected size (clamped by model context size) instead of always reporting the model max context. This will help provide a hint to clients to keep the context size below this value to avoid paging and poor performance on smaller VRAM systems.

    下载附件