Kimi K3 took ~500MB of 3D assets down to 17.8MB
Then generated 56 historical images and optimized those too. Kimi K3 is no doubt leading DeepSeek V4, Qwen 3.8 Max, Opus 5, and GPT-5.6 Over 5 hours of work, but the result is seriously impressive. 🔥
DeepSeek V4 Pro (0813) vs Qwen 3.8 Max
GLM 5.3 Max vs Fable 5 vs Qwen 3.8 Max vs Grok 4.6
GLM 5.3 is a massive jump over 5.2. Still not beating Fable 5 for me, but it’s way closer now. And vs Grok 4.6? I’d take GLM 5.3 pretty easily on this test.
Advanced Frontier LLM Coding Benchmark 2 14.08.2026
Models: - Opus 5 Max - Claude Code (App) - Qwen 3.8 Max - Qwen-Code (CLI) - GLM 5.3 Max - Zcode (App) - GPT 5.6 Sol Ultra - Codex (App) Task: Ferrofluid: Rising towards metaball surface + magnet cursor
You have reached the end of the archive
All of qwen38