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） |
|---|---|---|---|---|---|---|---|
| #371 | Grok 4.1 Fast medium | X AI | 0.8 | 0.4 | $0.069 | 0/1 | 23.6s |
| #261 | Gemini 3.1 Flash Lite none | 0.5 | 0.6 | $0.069 | 1/3 | 938ms | |
| #356 | KAT Coder AIR V2.5 default | Kwaipilot | 0.3 | 0.4 | $0.070 | 0/3 | 14.3s |
| #341 | Qwen3 Coder Next medium | Qwen | 0.4 | 0.5 | $0.071 | 0/3 | 924ms |
| #122 | Qwen3.7 Flash low | Qwen | 0.7 | 0.8 | $0.071 | 1/3 | 42.4s |
| #151 | Qwen3.8 27B medium | Qwen | 1.0 | 0.7 | ~$0.073 | 3/3 | 40.5s |
| #394 | Command A+ low | Cohere | 0.3 | 0.3 | $0.073 | 0/3 | 86.4s |
| #118 | GPT-6 Luna high | OpenAI | 0.8 | 0.8 | $0.074 | 2/3 | 24.7s |
| #200 | Qwen3.7 Flash medium | Qwen | 0.6 | 0.7 | $0.076 | 1/3 | 46.8s |
| #309 | KAT Coder AIR V2.5 high | Kwaipilot | 0.4 | 0.5 | $0.077 | 0/3 | 39.1s |
| #230 | Solar Pro 4 high | Upstage | 0.6 | 0.6 | $0.077 | 1/3 | 244.7s |
| #390 | Command A+ high | Cohere | 0.3 | 0.3 | $0.079 | 0/3 | 92.7s |
| #259 | Gemini 3.1 Flash Lite minimal | 0.5 | 0.6 | $0.080 | 1/3 | 831ms | |
| #284 | Nemotron 3 Super medium | NVIDIA | 0.3 | 0.6 | $0.081 | 0/3 | 147.3s |
| #139 | DeepSeek V4 Pro 0423 none | DeepSeek | 0.6 | 0.7 | $0.083 | 1/3 | 13.4s |