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 ↓.
313/313
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） |
|---|---|---|---|---|---|---|---|
| #278 | Grok 4.20 Multi Agent Beta medium | X AI | 4 | 4.8 | $5.599 | 8/18 | 9.69s |
| #5 | GPT-6 Astra xhigh | OpenAI | 1 | 9.9 | $5.156 | 21/22 | 28.4s |
| #27 | GPT-5.5 medium | OpenAI | 4 | 9.0 | $4.163 | 18/22 | 39.0s |
| #4 | GPT-6 Astra high | OpenAI | 1 | 9.9 | $3.474 | 21/22 | 18.6s |
| #25 | Claude Fable 5.1 high | Anthropic | 3 | 9.0 | $3.364 | 19/22 | 13.6s |
| #20 | Claude Opus 5 high | Anthropic | 2 | 9.2 | $3.070 | 18/22 | 22.5s |
| #44 | Claude Fable 5 medium | Anthropic | 1 | 8.6 | $3.004 | 17/22 | 15.0s |
| #87 | Claude Opus 4.6 medium | Anthropic | 4 | 7.7 | $2.961 | 13/22 | 32.2s |
| #57 | Claude Fable 5.1 medium | Anthropic | 3 | 8.3 | $2.909 | 15/22 | 11.7s |
| #61 | Qwen3.8 2.4T A95B low | Qwen | 4 | 8.1 | $2.672 | 15/22 | 119.0s |
| #141 | Claude Fable 5.1 low | Anthropic | 4 | 6.9 | $2.565 | 11/22 | 10.3s |
| #110 | Qwen3.8 2.4T A95B high | Qwen | 3 | 7.4 | $2.357 | 14/22 | 120.6s |
| #250 | Gemini 3.1 Flash Lite Preview high | 2 | 5.3 | $2.310 | 13/16 | 68.1s | |
| #65 | Muse Spark 1.1 high | Meta | 3 | 8.0 | $2.253 | 12/22 | 36.9s |
| #29 | Grok 4.5 high | X AI | 2 | 9.0 | $2.252 | 17/22 | 72.3s |