Four models, same scene.

Grok 4.6 finished last, behind Opus 5, Qwen 3.8 Max and GLM 5.3 Max. look at the clocks though. Grok finished in roughly 15 minutes. Opus took about 45. three times faster and fourth on looks. in a one-shot beauty contest that reads as a loss, and it is

88n77Aug 153

Qwen 3.8 Max Full COURSE 1 HOUR (Build & Automate Anything)

Julian Goldie SEOAug 151

The secret behind Qwen 3.8 27B is GRPO distillation using Qwen 3.8 Max outputs as the RL objective

Letting the smaller model reason as much as it needs to match the larger one.

0xMarioNawfalAug 1531

Qwen 3.8 27b nvfp4 vs Grok 4.6

Someone tell me how this local model that is 100x smaller and 10x slower made a better animation?! elonmusk help us solve this, both with their highest thinking mode (xhigh vs max)

WëscheAug 1513