MiniMax (official)· @MiniMax_AI · X·· 2026-06-13AI 评分47
AI 导读
MiniMax M3 支持已添加到 mlx-vlm,包含 MSA 实现!🚀 在 M3 Ultra 512GB 上测试,运行速度 24 tps,峰值内存约 240GB。目前正在优化性能并添加大量测试 💪 模型:https://huggingface.co/mlx-community/MiniMax-M3-4bit PR:https://github.com/Blaizzy/mlx-vlm/pull/1374 我们喜爱社区用 M3 开放权重构建的一切 ♥️ 期待下一步
正文 · 原文
We love what the community is building with M3 open weights ♥️
Excited for what’s next
MiniMax M3 support added to mlx-vlm with MSA implementation! 🚀 Tested on M3 Ultra 512GB running at 24 tps with peak memory ~240GB. Now working on optimizing performance and adding ton of tests 💪 Model is here: https://huggingface.co/mlx-community/MiniMax-M3-4bit PR is here: https://github.com/Blaizzy/mlx-vlm/pull/1374在 X 查看被引用的帖子
来源:MiniMax (official) · x.com