提供免费的、兼容OpenAI的API接口，支持多个开源大语言模型，无需注册即可使用。
Performance Stats
Avg Speed
38.73t/s
Latency
1.60s
Total Tests
9,136
Models
49
| Model | Speed | Latency | Tests |
|---|---|---|---|
| deepseek/deepseek-v3 | 172.09 t/s | 2.64s | 90 |
| zhipu/glm-4.6v-flash | 145.99 t/s | 11.14s | 10 |
| zhipu/glm-4.6v-flash | 145.99 t/s | 11.14s | 10 |
| zhipu/glm-4.6v-flash | 145.99 t/s | 11.14s | 10 |
| deepseek/deepseek-v3-0324 | 127.92 t/s | 3.06s | 520 |
| zhipu/glm-4.1v-thinking-flash | 105.37 t/s | 7.62s | 50 |
| zhipu/glm-4.1v-thinking-flash | 105.37 t/s | 7.62s | 50 |
| xiaomi/mimo-v2-flash | 104.74 t/s | 2.61s | 20 |
| openai/gpt-oss-20b | 87.74 t/s | 1.97s | 150 |
| qwen/qwen2.5-7b | 80.97 t/s | 1.76s | 95 |
| qwen/qwen2.5-7b | 80.97 t/s | 1.76s | 95 |
| zhipu/glm-4-9b | 64.28 t/s | 1.05s | 15 |
| zhipu/glm-4v-flash | 56.78 t/s | 0.87s | 50 |
| google/gemma-3-27b | 41.19 t/s | 2.38s | 190 |
| google/gemini-2.0-flash-exp | 36.13 t/s | 0.71s | 355 |
| qwen/qwen3-30b-a3b | 34.89 t/s | 4.09s | 175 |
| zhipu/glm-4-flash | 34.75 t/s | 0.94s | 5280 |
| qwen/qwen3-8b | 31.68 t/s | 1.44s | 5 |
| meituan/longcat-flash-chat | 30.03 t/s | 2.06s | 30 |
| zhipu/glm-4.5-flash | 28.75 t/s | 14.27s | 80 |
| Time | Model | Speed | Latency |
|---|---|---|---|
| Jan 18, 10:50 AM | zhipu/glm-4-flash | 31.14 t/s | 2.12s |
| Jan 18, 09:42 AM | qwen/qwen3-8b | 31.68 t/s | 1.44s |
| Jan 18, 09:41 AM | qwen/qwen2.5-72b | 0.00 t/s | 0.00s |
| Jan 18, 09:41 AM | deepseek/deepseek-v3-0324 | 0.00 t/s | 0.00s |
| Jan 17, 04:41 PM | zhipu/glm-4.6v-flash | 134.33 t/s | 12.98s |
| Jan 17, 04:41 PM | qwen/qwen3-coder | 0.00 t/s | 2.14s |
| Jan 17, 04:38 PM | zhipu/glm-4-flash | 28.77 t/s | 1.21s |
| Jan 17, 04:24 PM | zhipu/glm-4-flash | 28.95 t/s | 1.65s |
| Jan 17, 06:35 AM | zhipu/glm-4v-flash | 55.76 t/s | 2.95s |
| Jan 17, 06:35 AM | qwen/qwen2.5-vl-32b | 0.00 t/s | 0.00s |