AI BENCHY 失败分析
答案错误 失败
看看哪些 AI 模型最常遇到 答案错误，让你在选择前先发现稳定性风险。 排序方式: 分数 ↓.
| 排名 | 模型 | 公司 | 答案错误 次数 | 分数 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|
| #17 | Gemini 3.1 Flash Lite Preview medium | 4 | 8.2 | 13/18 | 3.74s | |
| #18 | GLM 5 Turbo medium | Z.ai | 3 | 8.1 | 12/18 | 17.7s |
| #19 | Qwen3.5-122B-A10B medium | Qwen | 3 | 8.1 | 13/18 | 31.4s |
| #20 | Qwen3.6 Plus medium | Qwen | 3 | 8.1 | 13/18 | 15.3s |
| #21 | Gemini 3 Flash Preview none | 5 | 8.1 | 13/18 | 1.65s | |
| #22 | Gemini 3.1 Flash Lite Preview low | 4 | 8.1 | 13/18 | 3.22s | |
| #23 | MiMo-V2-Pro medium | Xiaomi | 3 | 8.1 | 12/18 | 12.3s |
| #24 | Gemma 4 26B A4B medium | 2 | 8.0 | 13/18 | 25.0s | |
| #25 | Grok 4.20 Beta medium | X AI | 3 | 8.0 | 12/18 | 9.81s |
| #26 | Claude Sonnet 4.6 medium | Anthropic | 2 | 8.0 | 13/18 | 12.7s |
| #27 | DeepSeek V3.2 medium | DeepSeek | 3 | 8.0 | 12/18 | 46.4s |
| #28 | GPT-5.2 Chat none | OpenAI | 5 | 7.9 | 12/18 | 6.84s |
| #29 | Gemini 3.1 Flash Lite Preview none | 4 | 7.9 | 12/18 | 1.30s | |
| #30 | Step 3.5 Flash medium | Stepfun | 3 | 7.9 | 11/17 | 26.8s |
| #31 | GLM 5V Turbo medium | Z.ai | 3 | 7.8 | 11/18 | 15.0s |