Ollama v0.32.10 发布：默认重复惩罚调整与性能优化

AI 摘要

Ollama 发布 v0.32.10 版本，主要更改包括：未设置 repeat_penalty 的模型默认值从 1.1 改为 1.0（关闭），以匹配其他引擎并加速推测解码；NVFP4 MLX 模型全局缩放使预填充速度提升约 7-8%；修复了 OCI 清单配置和层共享摘要时 blob 验证被跳过的问题。

正文节选

## What's Changed - Models that don't set a `repeat_penalty` now default to 1.0 (off) instead of 1.1, matching other engines and speeding up speculative decoding; set a per-model parameter if an older model repeats itself. - Faster prefill on NVFP4 MLX models with a global scale, about 7–8% on Qwen3.6 and Muse Glimmer. - Fixed blob verification being skipped when an OCI manifest's config and layer share a digest. ## New Contributors * @vigneshakaviki made their first contribution in https

发布时间：2026-08-13 06:36

抓取时间：2026-08-13 07:08

来源机构：Ollama