416 tok/s on a 3090. That is what promises for Qwen 3.6 35B-A3B, and it is the one column in that tool you should ignore.
The tool went round yesterday and it deserves the attention. Free, no signup, reads your GPU straight from the browser, 284 devices
In today’s mainstream model, creativity is no longer a shortcoming. Gemini 3.7 Flash, Grok 4.6
In today’s mainstream model, creativity is no longer a shortcoming. Gemini 3.7 Flash, Grok 4.6, Qwen 3.8 Max, Claude Opus 5, the same Ace of Spades, required to be drawn into the card. This kind of same-question test can better see the differences in models than running scores. The question is not who draws better, but who thinks more like a human being. 🔥…
Qwen 3.8 27B vs Fable, Kimi K3, GPT 5.6 Sol Pro on Flappy Bird
ModelScope just put a countdown on Qwen3.8-Flash-Next.
No weights yet. Card says multimodal MoE, 125B total, 6B active, plus a 51B n-gram embedding. GDN + QSA. They call it the Qwen4 architecture, shipping early so people can prep. I already run Qwen 3.8-27B on Spark 1. This is
You have reached the end of the archive
All of qwen38