AI BENCHY Category
Puzzle Solving Ranking
See which AI models perform best on Puzzle Solving, which ones stay reliable, and where the biggest gaps appear. Sort by: Tests Correct ↑.
| Rank | Model | Company | Puzzle Solving Score | Score | Tests Correct | Response Time (avg) |
|---|---|---|---|---|---|---|
| #62 | Gemini 2.5 Flash none | 5.7 | 6.2 | 1/3 | 576ms | |
| #65 | MiMo-V2-Pro none | Xiaomi | 6.0 | 6.0 | 1/3 | 1.83s |
| #66 | GPT-5.4 none | OpenAI | 5.6 | 5.9 | 1/3 | 1.52s |
| #67 | Qwen3.5-27B none | Qwen | 6.7 | 5.9 | 1/3 | 1.37s |
| #70 | Qwen3.5-122B-A10B none | Qwen | 5.4 | 5.7 | 1/3 | 982ms |
| #71 | MiniMax M2.5 medium | Minimax | 5.3 | 5.7 | 1/3 | 11.5s |
| #72 | Hunter Alpha none | OpenRouter | 5.8 | 5.7 | 1/3 | 3.06s |
| #75 | GLM 5.1 none | Z.ai | 5.7 | 5.6 | 1/3 | 1.48s |
| #77 | GLM 5 Turbo none | Z.ai | 5.5 | 5.5 | 1/3 | 2.43s |
| #78 | Trinity Large Preview none | Arcee AI | 5.4 | 5.3 | 1/3 | 3.30s |
| #79 | Grok 4.20 Beta none | X AI | 5.9 | 5.3 | 1/3 | 541ms |
| #82 | Grok 4.20 none | X AI | 5.3 | 5.2 | 1/3 | 487ms |
| #86 | GPT-5.4 Mini none | OpenAI | 5.4 | 5.1 | 1/3 | 860ms |
| #88 | Nemotron 3 Super none | NVIDIA | 5.7 | 5.1 | 1/3 | 7.50s |
| #6 | Seed-2.0-Lite medium | Bytedance Seed | 9.0 | 8.6 | 2/3 | 11.0s |