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: 总成本 ↓.
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） |
|---|---|---|---|---|---|---|---|
| #274 | Mistral Large 4 none | Mistral | 0.5 | 0.6 | $0.290 | 0/3 | 85.0s |
| #123 | Muse Glimmer 30B medium | Meta | 0.8 | 0.8 | $0.288 | 2/3 | 33.7s |
| #17 | Gemini 3.7 Flash medium | 0.8 | 1.0 | $0.278 | 2/3 | 8.86s | |
| #90 | Seed-2.0-Lite medium | Bytedance Seed | 0.8 | 0.8 | $0.271 | 2/3 | 156.7s |
| #302 | Kimi K2.5 none | Moonshot AI | 0.5 | 0.5 | $0.262 | 1/3 | 24.6s |
| #94 | DeepSeek V4 Flash 0423 high | DeepSeek | 0.8 | 0.8 | $0.261 | 2/3 | 50.6s |
| #291 | Seed-2.0-Code none | Bytedance Seed | 0.4 | 0.5 | $0.257 | 0/3 | 13.4s |
| #172 | Gemini 3 Flash Preview low | 0.6 | 0.7 | $0.254 | 1/3 | 6.00s | |
| #199 | GLM 5.2 none | Z.ai | 0.4 | 0.7 | $0.250 | 0/3 | 7.55s |
| #98 | GPT-5.4 Nano medium | OpenAI | 0.6 | 0.8 | $0.235 | 1/3 | 19.1s |
| #303 | Qwen3.6 Max Preview none | Qwen | 0.4 | 0.5 | $0.228 | 0/3 | 3.12s |
| #173 | Muse Glimmer 30B low | Meta | 0.8 | 0.7 | $0.219 | 2/3 | 16.5s |
| #109 | MiMo-V2.6-Flash medium | Xiaomi | 0.7 | 0.8 | $0.218 | 1/3 | 18.2s |
| #249 | Qwen3.5 Plus 2026-04-20 none | Qwen | 0.4 | 0.6 | $0.216 | 0/3 | 1.69s |
| #209 | GLM 5.3 low | Z.ai | 0.6 | 0.6 | $0.215 | 1/3 | 21.0s |