AI BENCHY 分类失败
编程：答案错误
编程
答案错误
看看哪些 AI 模型在 编程 上最容易遇到 答案错误，更快找出薄弱点。 排序方式: 响应时间（平均） ↓.
| 排名 | 模型 | 公司 | 答案错误 次数 | 分类得分 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|
| #68 | Seed-2.0-Mini medium | Bytedance Seed | 1 | 6.8 | 1/2 | 220.5s |
| #33 | Qwen3.6 Plus medium | Qwen | 1 | 4.1 | 0/2 | 201.7s |
| #20 | Qwen3.5 Plus 2026-02-15 medium | Qwen | 1 | 7.6 | 1/2 | 193.8s |
| #71 | DeepSeek V3.2 medium | DeepSeek | 1 | 3.9 | 0/2 | 185.0s |
| #72 | MiMo-V2-Omni medium | Xiaomi | 1 | 3.4 | 0/2 | 183.9s |
| #14 | Qwen3.6 Max Preview medium | Qwen | 1 | 8.2 | 1/2 | 178.0s |
| #42 | Qwen3.5 Plus 2026-04-20 medium | Qwen | 1 | 5.4 | 1/2 | 137.5s |
| #119 | MiniMax M2.5 medium | Minimax | 1 | 3.5 | 0/2 | 125.8s |
| #25 | Qwen3.5-27B medium | Qwen | 1 | 7.0 | 1/2 | 123.9s |
| #38 | Qwen3.5-122B-A10B medium | Qwen | 1 | 4.1 | 0/2 | 119.6s |
| #21 | Seed-2.0-Lite medium | Bytedance Seed | 1 | 7.0 | 1/2 | 107.7s |
| #1 | Gemini 3 Flash Preview medium | 1 | 7.9 | 1/2 | 96.0s | |
| #44 | MiMo-V2-Pro medium | Xiaomi | 1 | 7.5 | 1/2 | 94.2s |
| #105 | Cobuddy medium | Baidu | 1 | 4.1 | 0/2 | 79.2s |
| #65 | GPT-5.4 Mini medium | OpenAI | 1 | 7.5 | 1/2 | 73.3s |