Hey gang, here is my full evaluation/comparison of Qwen 3.8 27B against everyone's favorite, Qwen 3.6 27B.
I also found the first case of NVFP4 precision loss against BF16 that I could actually demonstrate. There was really no way I can do the full comparison justice in single
I tested four frontier models with the exact same prompt
Gemini 3.7 Flash, Qwen 3.8, Grok 4.6, and GLM 5.3. The task included two scenes: a working harvester and farmers in the field. Honestly, Qwen 3.8 impressed me the most—it simply understood the assignment and
Qwen 3.8 is out and you can now connect it to Telegram using Nemoclaw!
See how to set up your AI assistant on your phone and chat with it anytime. The possibilities are endless!
Roughly around 40 minutes to build a Pacman game with Qwen 3.8 27B.
It added sfx in it .. how is this possible. 🤯
You have reached the end of the archive
All of qwen38