AI Coding
Walk training, three models — Claude Fable 5.2 vs Qwen4-Max vs Gemini 4 Hydrogen
Scoreboard3
- Claude Fable 5.2Anthropic
- Qwen4-MaxAlibaba (Qwen)
- Gemini 4 HydrogenGoogle
Prompt
Walk training, three modelsThe post doesn’t quote the prompt wording.
The author uses the same prompt, with neural networks and a genetic algorithm, so Claude Fable 5.2, Qwen4-Max and Gemini 4 Hydrogen learn to walk. The author says Gemini beat Fable by 10 cm but on its knees, and that only Fable 5.2 actually walks.
The post, itemised
- Kind
- Side-by-side
- Task
- Walk training, three models
- Category
- Simulations
- Mode
- Not stated
- Posted
- Video length
- 58.5 s
- Likes
- 69
- Views
- 14,901
- Bookmarks
- 31
- Sponsored
- No
Related
Posted by @noclipepe on X. aivs.tv claims no ownership: it hosts a copy for fast playback and links to the post on X. Request removal


