← 返回列表

@danielhanchen: Unsloth Studio now has auto speculative decoding & MTP support for GGUFs! Get up to 2x faster inference with no accuracy loss! We ran m...

@danielhanchen 3 信息等级 3 1 噪音/剔除;2 较弱;3 普通事实;4 重要行业动态;5 极重大事件。该分数是信息显著性,不是投资建议。 发布:2026-05-19T16:41 抓取:2026-05-19 23:19
🔗 原文链接
摘要

Unsloth Studio 推出新功能,支持自动推测解码和 MTP,可将推理速度提升至 2 倍且无精度损失,并针对 Mac、GPU 和 CPU 优化了参数。

客观事实
  • Unsloth Studio 新增自动推测解码和 MTP 支持
  • 推理速度提升可达 2 倍且无精度损失
  • 已针对 Mac、GPU 和 CPU 优化参数
Unsloth Studio

原文

Unsloth Studio now has auto speculative decoding & MTP support for GGUFs! Get up to 2x faster inference with no accuracy loss!

We ran many experiments from small models to MoEs, and optimized the params for Mac, GPUs & CPUs.

There's also a new toggle for MTP / ngram or auto!

likes: 101 | retweets: 11 | replies: 5 | views: 8783