Ornith 1.5 397B Max + Qwen 3.8 27B Vision = AGI at home
Playing with Ornith 1.5 35B like crazy ~30M tokens just today with oMLX!
This model is super fast locally on Apple Silicon being a MoE! I still have to test it on DGX Spark. Here I'm testing it as coder following an ultra-detailed plan defined by Kimi K3 and reviewed by GLM
Gemini 3.7 vs Qwen 3.8 Max vs Kimi K3
I gave all three models the exact same one-line prompt. Results: • Kimi K3 → ~8 min • 890 lines of code • Qwen 3.8 Max → ~12 min • 1,190 lines of code • Gemini 3.7 → ~2 min • 858 lines of code My take: • Kimi K3 → Excellent
Me running Qwen 3.8-27B on my machine 🚀
You have reached the end of the archive
All of qwen38