Qwen 3.8 max just beat GPT-5.6 sol max at actually using a computer.
But 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 for huge SOPs,
90 rotations of 15 minutes with Qwen 3.8 27B on my DGX Spark (Asus Ascent GX10).
That’s 22.5 hours of pure iteration. The result: this complete LEGO New York City skyline 🗽🏙️
Gemini 3.7 Flash High vs Claude Opus 5 High vs Grok 4.6 High vs Qwen 3.8 27B
You have reached the end of the archive
All of qwen38