Video by noclipepe (@noclipepe) · originally posted on XView original post on X
Claude Haiku 5.5 vs GPT-6 Luna vs Grok 4.7
Cloth simulation, 4 to 4,096 faces
noclipepe@noclipepe82 likes16,983 views
The launch-day cloth prompt on Claude Haiku 5.5, GPT-6 Luna and Grok 4.7, with times and tokens as stated. The author found Haiku's folds most convincing.
Models
| Model | Time | Tokens |
|---|---|---|
| Claude Haiku 5.5Anthropic | 33 min | 238.7K tokens |
| GPT-6 LunaOpenAI | 13 min | 47.5K tokens |
| Grok 4.7xAI | 1 h 36 min | 457.8K tokens |
Details
| Kind | Side-by-side (same post, same prompt) |
|---|---|
| Task | Cloth simulation, 4 to 4,096 faces |
| Author’s verdict | Haiku 5.5 gave the most convincing folds and draping; Luna finished first. — @noclipepe’s view (paraphrased) |
| Topic | Interactive cloth simulation |
| Prompt (from the post) | build an interactive cloth simulation, from 4 to 4,096 faces |
| Category | Simulations |
| Mode | Not stated |
| Posted | |
| Video length | 35.6 s |
| Likes / views | 82 / 16,983 · as of Oct 18, 2026 |
| Bookmarks | 40 |
| Sponsored | No |
| Original post | x.com/noclipepe/status/2107974894667333674 |
Related
Compare all 3 modelsClaude Haiku 5.5 vs GPT-6 LunaClaude Haiku 5.5 vs Grok 4.7GPT-6 Luna vs Grok 4.7
Content belongs to its creators. Video by @noclipepe; aivs.tv hosts a copy for fast playback and links to the original post on X. Request removal
