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 ↑.
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) |
|---|---|---|---|---|---|---|---|
| #303 | Grok 4.1 Fast none | X AI | 3.0 | 3.8 | $0.008 | 0/3 | 1.10s |
| #304 | Qwen3.5-9B medium | Qwen | 3.0 | 3.8 | $0.062 | 0/3 | 32.3s |
| #305 | Ling 3.0 Tiny medium | Inclusionai | 3.6 | 3.8 | $0.000 | 0/3 | 34.7s |
| #308 | Nemotron 3 Nano Omni 30b A3b Reasoning medium | NVIDIA | 2.9 | 3.4 | $0.000 | 0/3 | 1.40s |
| #309 | Nemotron 3 Nano Omni 30b A3b Reasoning none | NVIDIA | 3.0 | 3.2 | $0.000 | 0/3 | 532ms |
| #310 | Step 3.5 Flash none | Stepfun | 0.0 | 2.3 | $0.020 | 0/0 | 0ms |
| #311 | LFM2-24B-A2B none | Liquid | 3.8 | 2.2 | $0.001 | 0/3 | 1.78s |
| #57 | GPT-5 Mini medium | OpenAI | 5.6 | 8.1 | $0.241 | 1/3 | 15.2s |
| #59 | Inkling high | Thinkingmachines | 6.9 | 8.0 | $1.046 | 1/3 | 10.7s |
| #63 | GLM 5.2 high | Z.ai | 6.0 | 8.0 | $1.188 | 1/3 | 33.7s |
| #68 | Step 3.7 Flash medium | Stepfun | 5.7 | 7.9 | $0.495 | 1/3 | 6.19s |
| #79 | Qwen3.7 Flash high | Qwen | 7.1 | 7.8 | $0.052 | 1/3 | 10.8s |
| #83 | DeepSeek V4 Pro high | DeepSeek | 6.9 | 7.7 | $0.489 | 1/3 | 56.8s |
| #88 | Nemotron 3 Ultra medium | NVIDIA | 5.5 | 7.6 | $0.567 | 1/3 | 3.54s |
| #102 | Kimi K2.7 Code medium | Moonshot AI | 5.9 | 7.4 | $0.687 | 1/3 | 41.0s |