Qwen 3.8 just made everyone else look like they missed the prompt.

Bhavani_00007 tested Qwen 3.8, GLM 5.3, Grok 4.6, and Gemini 3.7 Flash on the exact same two-scene prompt. Qwen simply did what was asked. GLM came close. Grok messed up the coloring. Gemini fell behind. Qwen

noclipepeAug 165

This is QWEN 3.8 27B running Hyper(His name is Core "PID1" the god process at ~100/133 tok/s

I had it build me a react to do app, and deploy it, without me touching anything at all. Just talkin' to my AI

Rob McElvennyAug 161

Qwen-3.8-27b

I tried creating a virtual aquarium using Three.js. The background music was only added in video editing, but is actually silent. If you can prepare your own 3D model, you can do a lot of things. BGM used Fluttering water: Takayuki Amanuma / Audiostock

ほりんち🍓Aug 162

Got myself a local Qwen 3.8 provider with my son's gaming laptop (nvidia 5090)

Decent speed. Free compute while he's in school. Supposedly Qwen 3.8 outperforms Grok 4.6 in real life tasks.

GUBAAug 161