Qwen 3.8 on amd GPU

KelvinNjiroAug 14

Qwen 3.8 - 27B

It was finally released so I tried it right away. Is it the setting that is unusually slow (9t/s)? Is the default size (8bit quantization model) mismatched to 128GB/m5max? In any case, I quickly shared the data as it was interesting. 1st LMstudio Simple Reasoning 2nd ollama+claudecode collaboration inference large…

佐藤 勝彦（TANREN_CEO)┃生成AIエバンジェリストAug 141

I can create a creative agent using OpenCode, MiniMax-H3, and Qwen 3.8 27B, but how do I manage VRAM?

あるふAug 1423

9 days ago Qwen's blog said "next week."

I probed the repo every single morning. 404. 404. 404. This morning: Qwen 3.8 27B. Apache 2.0. Inside the window. And the headline everyone's running is the LEAST interesting thing about it. 🧵

Signal CodersAug 14