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。
| 排名 | 模型 | 公司 | 答案错误 次数 | 分数 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|
| #41 | Nemotron 3 Ultra 550b A55b medium | NVIDIA | 7 | 7.5 | 13/21 | 15.1s |
| #44 | Gemini 3.1 Flash Lite medium | 7 | 7.5 | 13/21 | 3.23s | |
| #50 | Gemini 3.1 Flash Lite Preview low | 7 | 7.4 | 13/21 | 2.77s | |
| #58 | Gemini 3.1 Flash Lite Preview none | 7 | 7.2 | 12/21 | 1.21s | |
| #59 | GLM 5V Turbo medium | Z.ai | 7 | 7.2 | 11/21 | 23.1s |
| #63 | GPT-5.3 Chat none | OpenAI | 7 | 7.2 | 12/21 | 6.34s |
| #100 | Grok Build 0.1 none | X AI | 7 | 6.0 | 7/19 | 28.7s |
| #129 | MiniMax M2.5 medium | Minimax | 7 | 5.3 | 5/21 | 65.4s |
| #133 | DeepSeek V3.2 none | DeepSeek | 7 | 5.2 | 6/21 | 13.8s |
| #149 | Nemotron 3 Nano Omni 30b A3b Reasoning medium | NVIDIA | 7 | 4.6 | 4/19 | 17.1s |
| #24 | GPT-5.2 Chat none | OpenAI | 6 | 7.9 | 14/21 | 7.13s |
| #28 | Gemini 2.5 Flash medium | 6 | 7.8 | 14/21 | 15.5s | |
| #31 | DeepSeek V4 Flash high | DeepSeek | 6 | 7.7 | 13/21 | 45.8s |
| #45 | GPT-5.4 Mini medium | OpenAI | 6 | 7.5 | 12/21 | 22.3s |
| #65 | Grok 4.20 medium | X AI | 6 | 7.1 | 12/21 | 27.7s |