AI BENCHY Category Failures
Domain specific: Wrong answer
Domain specific
Wrong answer
See which AI models are most likely to hit Wrong answer on Domain specific, so you can spot weak points faster. Sort by: Tests Correct ↓.
Failure Reasons
| Rank | Model | Company | Wrong answer Count | Category Score | Tests Correct | Response Time (avg) |
|---|---|---|---|---|---|---|
| #69 | Claude Opus 4.6 medium | Anthropic | 1 | 3.0 | 0/3 | 83.4s |
| #71 | Step 3.7 Flash high | Stepfun | 2 | 4.1 | 0/3 | 149.6s |
| #72 | DeepSeek V3.2 medium | DeepSeek | 2 | 2.9 | 0/3 | 24.3s |
| #75 | Ring-2.6-1T medium | Inclusionai | 3 | 3.5 | 0/3 | 64.9s |
| #76 | Kimi K2.5 medium | Moonshot AI | 2 | 3.5 | 0/3 | 137.3s |
| #78 | Qwen3.6 27B medium | Qwen | 3 | 2.9 | 0/3 | 73.4s |
| #79 | Hunter Alpha medium | OpenRouter | 1 | 3.0 | 0/3 | 10.5s |
| #80 | Mimo V2 Omni medium | Xiaomi | 1 | 3.0 | 0/3 | 47.9s |
| #81 | Mercury 2 medium | Inception | 3 | 2.9 | 0/3 | 6.48s |
| #84 | Grok 4.20 Multi Agent Beta medium | X AI | 2 | 2.9 | 0/3 | 24.7s |
| #87 | Gemini 3.1 Flash Lite minimal | 3 | 2.9 | 0/3 | 1.02s | |
| #88 | Qwen3.7 Plus none | Qwen | 3 | 3.0 | 0/3 | 868ms |
| #90 | Gemini 3.1 Flash Lite none | 3 | 2.9 | 0/3 | 762ms | |
| #91 | GPT-5.5 none | OpenAI | 3 | 2.9 | 0/3 | 1.31s |
| #93 | Qwen3.6 Plus Preview medium | Qwen | 2 | 3.0 | 0/3 | 22.1s |