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 ↑.
451/451
Filter models
No models match the current search and filters.
| Rank | Model | Company | Puzzle Solving Score | Score | Total Cost | Tests Correct | Response Time (avg) |
|---|---|---|---|---|---|---|---|
| #114 | Qwen3.8 Flash Next low | Qwen | 0.9 | 0.8 | ~$0.045 | 2/3 | 1.72s |
| #116 | Qwen3.5-27B medium | Qwen | 0.8 | 0.8 | $1.135 | 2/3 | 59.6s |
| #117 | GLM 5.3 FlashX max | Z.ai | 0.8 | 0.8 | $0.623 | 2/3 | 3.88s |
| #119 | GPT-5.4 Mini medium | OpenAI | 0.8 | 0.8 | $1.102 | 2/3 | 4.37s |
| #121 | Gemini 3.5 Flash medium | 0.8 | 0.8 | $1.316 | 2/3 | 2.38s | |
| #123 | MiMo-V2.6-Flash medium | Xiaomi | 0.8 | 0.8 | $0.218 | 2/3 | 52.9s |
| #124 | Muse Wonder max | Meta | 0.8 | 0.8 | $4.766 | 2/3 | 8.97s |
| #128 | GLM 5.3 high | Z.ai | 0.8 | 0.8 | $0.495 | 2/3 | 4.74s |
| #130 | Grok 4.6 medium | X AI | 0.9 | 0.8 | $2.270 | 2/3 | 16.3s |
| #131 | Qwen3.7 Plus none | Qwen | 0.8 | 0.8 | $0.178 | 2/3 | 1.71s |
| #133 | DeepSeek V4 Flash 0731 medium | DeepSeek | 0.8 | 0.8 | $0.460 | 2/3 | 65.1s |
| #135 | GPT-6 Luna high | OpenAI | 0.8 | 0.8 | $0.074 | 2/3 | 7.51s |
| #137 | Nemotron 3.5 Super VL medium | NVIDIA | 0.8 | 0.8 | $0.437 | 2/3 | 11.8s |
| #138 | Claude Haiku 5.5 medium | Anthropic | 0.8 | 0.8 | $0.063 | 2/3 | 2.67s |
| #139 | Qwen3.5 Plus 2026-04-20 medium | Qwen | 0.8 | 0.8 | $0.435 | 2/3 | 17.7s |