Here’s my first test of Qwen 3.8 27B on the inco_ai Splash inference engine.
✅ More than doubled the tokens/s that the M5 Pro was getting on oMLX. ✅ Two command setup ✅ Equal Game/Code quality
Bottom line! It’s hard to build a company and have big IPO’s when open weight models provide IP security and a less expensive solution.
Grok rank Qwen 3.8 vs common cloud models.
One AI model now watches and listens to your whole video
Qwen 3.8 Omni Flash turns it into a recap without chopping anything up.
Here's an example of Jev routing tools, skills, cli commands for Qwen 3.8 27B on cerebras to format a Power BI report.
Changes are almost instant; working on Linux, only wait due to publishing and refreshing embedded report. The video is not sped up, at all. Just a start.
You have reached the end of the archive
All of qwen38