指令遵循 排名
看看哪些 AI 模型在 指令遵循 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 ↑.
455/455
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 | 指令遵循 得分 | 分数 | 总成本 | 测试正确 | 响应时间（平均） |
|---|---|---|---|---|---|---|---|
| #261 | Solar Pro 4 high | Upstage | 0.8 | 0.6 | $0.077 | 1/2 | 41.5s |
| #426 | Command A+ high | Cohere | 0.3 | 0.3 | $0.079 | 0/2 | 110.3s |
| #291 | Gemini 3.1 Flash Lite minimal | 1.0 | 0.6 | $0.080 | 2/2 | 932ms | |
| #318 | Nemotron 3 Super medium | NVIDIA | 0.7 | 0.6 | $0.081 | 1/2 | 6.97s |
| #161 | DeepSeek V4 Pro 0423 none | DeepSeek | 0.6 | 0.7 | $0.083 | 1/2 | 4.12s |
| #240 | Laguna XS 2.1 medium | Poolside | 1.0 | 0.6 | $0.083 | 2/2 | 2.57s |
| #414 | Grok 4.20 Beta default | X AI | 0.6 | 0.4 | $0.087 | 1/2 | 649ms |
| #94 | Claude Haiku 5.5 high | Anthropic | 1.0 | 0.8 | $0.088 | 2/2 | 3.19s |
| #53 | GLM 5.3 Flash max | Z.ai | 1.0 | 0.9 | $0.089 | 2/2 | 3.65s |
| #148 | Qwen3.8 27B low | Qwen | 1.0 | 0.8 | ~$0.089 | 2/2 | 2.43s |
| #213 | Qwen3.5-Flash none | Qwen | 0.6 | 0.7 | $0.089 | 1/2 | 8.81s |
| #274 | Gemini 3.1 Flash Lite Preview none | 1.0 | 0.6 | $0.090 | 2/2 | 1.13s | |
| #352 | Laguna S 2.1 medium | Poolside | 0.4 | 0.5 | $0.092 | 0/2 | 400ms |
| #244 | Solar Pro 4 medium | Upstage | 1.0 | 0.6 | $0.098 | 2/2 | 40.1s |
| #214 | Solar Pro 4 low | Upstage | 0.8 | 0.7 | $0.101 | 1/2 | 56.4s |