刚在 oMLX 的社区跑分页面看到一组 M5 Ultra 数据，不知道有多真，但数字看起来相当合理。
This is a AI post classified by Jev as Open & local models (a model release), kept by the AI Radar because it carries real work, not commentary.
刚在 oMLX 的社区跑分页面看到一组 M5 Ultra 数据，不知道有多真，但数字看起来相当合理。 Qwen3.8 27B Q4 在 8K 上下文下，生成约 50 tok/s，Prefill 约 1800 tok/s，而且还没有开启 MTP。 如果数据属实，这已经是非常舒服的本地 Agent 速度。50 tok/s 足够实时写代码和调用工具，1800 tok/s 的 Prefill 也能快速读取大型项目和长文档。 后续加上 MTP、ANE Prefill 和更成熟的 oMLX 优化，生成速度还有继续提高的空间。
Posted by Wei (6.8k followers) 2 days ago · 17 likes · 5k views · view the original post on X. Kept by the AI Radar as Open & local models.
More AI work like this
- I'm done with Bonsai 2 27B. This model has taken enough of my time and patience. — @superalesha
- dear rtx 3060 owners, and every 12gb card behind it. bonsai 2 27b dense went from 26 to… — @sudoingX
- Nimble is able to process images! — @madiator
- Woah...it claims to be open Jev. And it's 100% opensource. — @Saboo_Shubham_
- your rtx 3060 was running bonsai 2 at 26 tok/s this morning and now does 40 tok/s, i… — @sudoingX
- Meet DeepSeek-V4.1-Flash: a blazing fast image-text-to-text model. It takes both images… — @HuggingModels
- Meet Edge0-35B-A3B-preview: a 35B parameter MoE model built for edge inference. It runs… — @HuggingModels
- Putting Jeff-1 out into the world, this is an attempt at a Jev like model, built on qwen… — @DJLougen
Every post is read and classified by Jev (TypeSafe): what it is, which market it belongs to, and whether the link is a real tool. 27.3k posts from 4.8k X accounts over the last 14 days, 1.2k tools, 19 markets. Collected every 5 minutes, fully re-ranked every hour — last update 2026-09-19 19:27 UTC. Full method.