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) |
|---|---|---|---|---|---|---|
| #59 | Qwen3.5-Flash none | Qwen | 3.3 | 6.2 | 0/3 | 5.90s |
| #63 | Qwen3.5-35B-A3B none | Qwen | 3.9 | 6.1 | 0/3 | 1.34s |
| #68 | gpt-oss-120b medium | OpenAI | 3.2 | 5.8 | 0/3 | 11.8s |
| #69 | Kimi K2.6 none | Moonshot AI | 3.4 | 5.8 | 0/3 | 1.66s |
| #73 | Mistral Small 4 medium | Mistral | 3.4 | 5.7 | 0/3 | 2.00s |
| #74 | GLM 4.7 Flash none | Z.ai | 4.4 | 5.6 | 0/3 | 1.00s |
| #76 | Kimi K2.5 none | Moonshot AI | 3.1 | 5.5 | 0/3 | 4.73s |
| #80 | MiniMax M2.7 medium | Minimax | 3.8 | 5.3 | 0/3 | 25.6s |
| #81 | Elephant medium | Openrouter | 3.7 | 5.2 | 0/3 | 867ms |
| #83 | Mistral Small 4 none | Mistral | 3.1 | 5.2 | 0/3 | 589ms |
| #84 | gpt-oss-120b none | OpenAI | 4.5 | 5.2 | 0/3 | 6.86s |
| #85 | Elephant none | Openrouter | 3.3 | 5.2 | 0/3 | 849ms |
| #87 | Qwen3 Coder Next none | Qwen | 3.2 | 5.1 | 0/3 | 22.9s |
| #89 | GPT-4o-mini none | OpenAI | 3.7 | 4.9 | 0/3 | 1.30s |
| #90 | Qwen3.5-9B none | Qwen | 3.2 | 4.8 | 0/3 | 683ms |