Puzzle Solving: Wrong answer
Puzzle Solving
Wrong answer
See which AI models are most likely to hit Wrong answer on Puzzle Solving, so you can spot weak points faster.
Failure Reasons
186/186
Filter models
No models match the current search and filters.
| Rank | Model | Company | Wrong answer Count | Category Score | Total Cost | Tests Correct | Response Time (avg) |
|---|---|---|---|---|---|---|---|
| #228 | Hy4 preview low | Tencent | 1 | 5.3 | $1.745 | 0/3 | 238.0s |
| #229 | Qwen3.6 27B none | Qwen | 1 | 5.3 | $0.116 | 1/3 | 5.15s |
| #230 | Hy3 preview low | Tencent | 1 | 5.3 | $0.042 | 1/3 | 7.51s |
| #231 | Qwen3.8 27B none | Qwen | 1 | 6.0 | ~$0.006 | 1/3 | 1.25s |
| #233 | DeepSeek V4 Flash 0423 none | DeepSeek | 1 | 3.1 | $0.037 | 0/3 | 23.7s |
| #236 | GPT-5.6 Luna none | OpenAI | 1 | 5.3 | $0.029 | 1/3 | 790ms |
| #237 | MiMo-V2.5-Pro none | Xiaomi | 1 | 6.7 | $0.068 | 1/3 | 1.30s |
| #246 | Qwen3.6 35B A3B none | Qwen | 1 | 3.2 | $0.051 | 0/3 | 1.07s |
| #248 | Inkling none | Thinkingmachines | 1 | 5.6 | $0.147 | 1/3 | 931ms |
| #249 | Granite 4.2 8B medium | IBM Granite | 1 | 3.1 | $0.009 | 0/3 | 76.0s |
| #255 | MiMo-V2.5 none | Xiaomi | 1 | 5.4 | $0.025 | 1/3 | 2.13s |
| #257 | GLM 5 Turbo none | Z.ai | 1 | 5.5 | $0.047 | 1/3 | 2.65s |
| #258 | North Mini Code none | Cohere | 1 | 3.5 | $0.000 | 0/3 | 24.4s |
| #259 | MiniMax M2.7 medium | Minimax | 1 | 5.9 | $0.208 | 1/3 | 24.9s |
| #260 | Mercury 2.5 Preview low | Inception | 1 | 8.2 | $0.011 | 2/3 | 710ms |