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: 总成本 ↑.
215/215
筛选模型
没有模型匹配当前搜索和筛选条件。
| 排名 | 模型 | 公司 | 答案错误 次数 | 分数 | 总成本 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|---|
| #56 | Kimi K2.7 Code medium | Moonshot AI | 5 | 7.5 | $0.740 | 12/22 | 84.2s |
| #3 | Gemini 3 Flash Preview medium | 1 | 9.6 | $0.742 | 21/22 | 19.2s | |
| #94 | Qwen3.6 35B A3B medium | Qwen | 4 | 6.7 | $0.746 | 13/22 | 58.1s |
| #137 | Grok 4.20 Beta medium | X AI | 3 | 6.0 | $0.750 | 14/18 | 9.75s |
| #60 | GPT-5.4 Mini medium | OpenAI | 6 | 7.5 | $0.756 | 12/22 | 25.9s |
| #55 | Nemotron 3 Ultra medium | NVIDIA | 7 | 7.5 | $0.774 | 13/22 | 32.2s |
| #79 | Grok 4.20 medium | X AI | 6 | 7.1 | $0.777 | 12/22 | 29.5s |
| #105 | Qwen3.6 27B medium | Qwen | 6 | 6.5 | $0.779 | 10/22 | 106.3s |
| #77 | Grok 4.3 medium | X AI | 5 | 7.1 | $0.779 | 13/22 | 47.4s |
| #35 | GLM 5.2 high | Z.ai | 3 | 8.0 | $0.817 | 14/22 | 62.7s |
| #1 | Gemini 3.6 Flash medium | 1 | 9.9 | $0.831 | 21/22 | 10.1s | |
| #125 | Qwen3.5-35B-A3B medium | Qwen | 2 | 6.2 | $0.837 | 11/22 | 112.5s |
| #16 | GPT-5.3-Codex medium | OpenAI | 4 | 8.9 | $0.920 | 16/22 | 17.0s |
| #26 | Claude Sonnet 5 medium | Anthropic | 4 | 8.3 | $0.922 | 16/22 | 12.5s |
| #23 | Grok 4.5 low | X AI | 6 | 8.4 | $0.935 | 16/22 | 15.6s |