I asked GPT 5.6 Luna to write a prompt for creating a website for a package delivery company and then fed that prompt into Grok 4.6 and Qwen 3.8
In the end, what I got wasn’t a website, but some kind of slideshow for children explaining how global delivery works Grok spent
Tools any normal person needs for generating excellent audio visual content is freely available now for home PC.
Fish Audio Pro/Chatterbox (Voice) Stable Audio 3 (SoundFx) Acestep 1.5 (Music) Krea 2 (Images) Ideogram4 (UI/layout) Minimax H3/LTX2.5 (Video) Qwen 3.8 (Intelligence)
Uncensored version of Qwen3.8-27B on 5090
I wanted to run it on NVFP4 with 200K context. The prefill seems fast and it seems easy to use. There were no ready-made ones, so I baked it myself. ▫️ 2 types in NVFP4 version All 4 bits, The MLP is 4 bits and the attention and output layers are 8 bits.
Qwen 3.8 max just beat GPT-5.6 sol max at actually using a computer.
And the benchmark isn't even the most interesting part. The numbers: → 2.4 trillion parameter Mixture-of-Experts model → Only ~95B parameters activate per token → 1M-token context window across text,
You have reached the end of the archive
All of qwen38