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. Sort by: Response Time (avg) ↑.
Failure Reasons
246/246
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) |
|---|---|---|---|---|---|---|---|
| #102 | GLM 5.2 medium | Z.ai | 1 | 0.8 | $1.007 | 2/3 | 13.1s |
| #217 | Apodex 1.1 Mini high | Apodex | 1 | 0.8 | $0.000 | 2/3 | 13.5s |
| #63 | DeepSeek V4.1 Flash high | DeepSeek | 1 | 0.6 | $0.617 | 1/3 | 14.3s |
| #157 | Inkling Small high | Thinkingmachines | 2 | 0.5 | $0.398 | 0/3 | 14.4s |
| #110 | GPT-5.6 Luna high | OpenAI | 1 | 0.8 | $0.118 | 2/3 | 14.6s |
| #282 | Apodex 1.1 Mini medium | Apodex | 1 | 0.8 | $0.000 | 2/3 | 15.2s |
| #84 | GPT-5 Mini medium | OpenAI | 1 | 0.6 | $0.315 | 1/3 | 15.2s |
| #421 | Xing4.0 29b A4b none | Xingchen AGI | 2 | 0.3 | $0.000 | 0/3 | 16.8s |
| #139 | Qwen3.5 Plus 2026-04-20 medium | Qwen | 1 | 0.8 | $0.435 | 2/3 | 17.7s |
| #42 | Pareto default | Unbiased | 1 | 0.8 | $1.520 | 2/3 | 18.2s |
| #209 | Grok Build 0.1 medium | X AI | 1 | 0.8 | $1.419 | 2/3 | 18.3s |
| #179 | Qwen3.8 Flash Next xhigh | Qwen | 1 | 0.8 | ~$0.195 | 2/3 | 19.5s |
| #325 | North Mini Code medium | Cohere | 3 | 0.3 | $0.000 | 0/3 | 19.7s |
| #302 | GPT-5 Nano medium | OpenAI | 1 | 0.5 | $0.159 | 1/3 | 20.6s |
| #300 | Ring 2.6 1t medium | Inclusionai | 1 | 0.6 | $0.102 | 1/3 | 20.7s |