GLM 5.3 had no business looking this good.
OmedTheVibeCoder tested GLM 5.3 Max against Fable 5, Qwen 3.8 Max, and Grok 4.6 on the same task. Fable 5 still takes it for me. But GLM 5.3 is suddenly much closer — and against Grok 4.6, I’d pick GLM pretty comfortably here. That’s
Qwen 3.8 27B by Alibaba_Qwen got released today aaaand here is
Qwen 3.8 27B vs. Qwen 3.6 27B vs Haiku 4.5 Prompt: Build flappy bird. 🟣 Qwen 3.8: thought for ~14 min before typing anything, then wrote the whole game in ~5 min. Total 19m 12s, 14,448 tokens. 🔵 Qwen 3.6:
SUPER TONY Coming Soon
Qwen 3.8 27B HTML Mario Inspired Game
Qwen 3.8 27b NVFP4
The same web prompts, not bad, but BF16 is definitely king.
You have reached the end of the archive
All of qwen38