Puzzle Solving Ranking
See which AI models perform best on Puzzle Solving, which ones stay reliable, and where the biggest gaps appear.
311/311
Filter models
No models match the current search and filters.
| Rank | Model | Company | Puzzle Solving Score | Score | Total Cost | Tests Correct | Response Time (avg) |
|---|---|---|---|---|---|---|---|
| #59 | Inkling high | Thinkingmachines | 6.9 | 8.0 | $1.046 | 1/3 | 10.7s |
| #83 | DeepSeek V4 Pro high | DeepSeek | 6.9 | 7.7 | $0.489 | 1/3 | 56.8s |
| #237 | MiMo-V2.5-Pro none | Xiaomi | 6.7 | 5.4 | $0.068 | 1/3 | 1.30s |
| #147 | MiMo-V2.5-Pro medium | Xiaomi | 6.7 | 6.8 | $0.221 | 1/3 | 5.31s |
| #158 | Qwen3.5-27B none | Qwen | 6.7 | 6.6 | $0.058 | 1/3 | 1.38s |
| #192 | Qwen3.5 Plus 2026-04-20 none | Qwen | 6.7 | 6.1 | $0.122 | 1/3 | 1.97s |
| #273 | Grok 4.20 Multi Agent Beta medium | X AI | 6.7 | 4.8 | $5.599 | 1/3 | 5.19s |
| #191 | Gemma 4 31B none | 6.5 | 6.1 | $0.016 | 1/3 | 4.23s | |
| #110 | KAT-Coder-Pro V2.5 low | Kwaipilot | 6.4 | 7.3 | $0.400 | 1/3 | 3.11s |
| #176 | Mimo V2 PRO medium | Xiaomi | 6.4 | 6.3 | $0.333 | 1/3 | 5.08s |
| #275 | GLM 4.7 Flash none | Z.ai | 6.4 | 4.8 | $0.016 | 1/3 | 1.20s |
| #188 | Inkling low | Thinkingmachines | 6.4 | 6.1 | $0.187 | 1/3 | 2.97s |
| #300 | Grok Build 0.1 none | X AI | 6.4 | 4.0 | $0.547 | 1/3 | 9.55s |
| #194 | Trinity Large Thinking low | Arcee AI | 6.3 | 6.0 | $0.625 | 1/3 | 3.28s |
| #186 | Gemini 3.1 Flash Lite none | 6.3 | 6.1 | $0.046 | 1/3 | 720ms |