编程 排名
看看哪些 AI 模型在 编程 上表现最好，哪些更稳定，以及差距主要出现在哪里。 排序方式: 测试正确 ↓.
408/408
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 | 编程 得分 | 分数 | 总成本 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|---|
| #29 | Claude Opus 5.5 medium | Anthropic | 0.8 | 0.9 | $1.804 | 2/3 | 13.5s |
| #30 | GPT-5.5 medium | OpenAI | 0.9 | 0.9 | $4.809 | 2/3 | 59.8s |
| #39 | Gemini 3.5 Flash low | 0.8 | 0.9 | $0.752 | 2/3 | 6.71s | |
| #40 | Qwen3.8 Max (0902) medium | Qwen | 0.8 | 0.9 | $1.096 | 2/3 | 45.7s |
| #42 | Gemini 3.5 Flash Lite high | 0.8 | 0.9 | $0.624 | 2/3 | 16.6s | |
| #43 | Claude Opus 5 low | Anthropic | 0.8 | 0.9 | $2.350 | 2/3 | 6.70s |
| #44 | Gemini 3.6 Flash low | 0.8 | 0.9 | $0.363 | 2/3 | 6.95s | |
| #50 | Qwen3.8 MAX low | Qwen | 0.8 | 0.9 | $1.089 | 2/3 | 28.7s |
| #51 | GPT-5.4 medium | OpenAI | 0.9 | 0.9 | $2.213 | 2/3 | 44.4s |
| #54 | Muse Spark 1.2 low | Meta | 0.8 | 0.9 | $0.881 | 2/3 | 10.1s |
| #56 | Qwen3.8 27B high | Qwen | 0.8 | 0.9 | ~$0.298 | 2/3 | 248.6s |
| #57 | MiMo-V2.6-Pro medium | Xiaomi | 0.8 | 0.9 | $0.586 | 2/3 | 150.4s |
| #58 | Seed 2.1 Turbo medium | Bytedance Seed | 0.8 | 0.9 | $2.042 | 2/3 | 354.9s |
| #62 | Claude Sonnet 5.5 xhigh | Anthropic | 0.8 | 0.9 | $1.516 | 2/3 | 11.9s |
| #64 | Claude Haiku 5.5 low | Anthropic | 0.8 | 0.9 | $0.061 | 2/3 | 7.30s |