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。
289/289
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） |
|---|---|---|---|---|---|---|---|
| #127 | Gemini 3.5 Flash none | 8.8 | 6.9 | $1.093 | 2/3 | 34.7s | |
| #5 | Gemini 3 Flash Preview medium | 8.6 | 9.5 | $0.763 | 2/3 | 84.4s | |
| #51 | Inkling high | Thinkingmachines | 8.5 | 8.0 | $1.041 | 2/3 | 148.4s |
| #56 | Inkling medium | Thinkingmachines | 8.4 | 8.0 | $0.377 | 2/3 | 25.2s |
| #110 | Muse Glimmer 30B medium | Meta | 8.4 | 7.2 | $0.280 | 2/3 | 33.7s |
| #7 | Gemini 3.7 Flash medium | 8.4 | 9.4 | $0.155 | 2/3 | 8.86s | |
| #25 | Gemini 3.5 Flash Lite high | 8.4 | 8.8 | $0.540 | 2/3 | 16.6s | |
| #35 | Qwen3.8 Max low | Qwen | 8.4 | 8.6 | $0.702 | 2/3 | 28.7s |
| #38 | Muse Spark 1.2 low | Meta | 8.4 | 8.4 | $0.655 | 2/3 | 10.1s |
| #80 | Nemotron 3 Ultra medium | NVIDIA | 8.4 | 7.6 | $0.780 | 2/3 | 26.5s |
| #88 | GPT-5.4 Mini medium | OpenAI | 8.4 | 7.5 | $0.786 | 2/3 | 57.9s |
| #284 | Laguna Xs.2 none | Poolside | 8.3 | 3.8 | $0.004 | 0/1 | 1.96s |
| #71 | Qwen3.7 Flash high | Qwen | 8.2 | 7.8 | $0.052 | 2/3 | 58.8s |
| #39 | Grok 4.6 medium | X AI | 8.2 | 8.4 | $1.713 | 2/3 | 120.0s |
| #43 | Seed 2.1 Turbo medium | Bytedance Seed | 8.2 | 8.4 | $1.951 | 2/3 | 354.9s |