智谱AI开放平台open.bigmodel.cn
0.43 s
Best: 0.34Worst: 1.10
40.94t/s
Leaderboard
Multi-dimensional rankings based on model speed tests and provider health checks. Compare providers, endpoints, and reliability at a glance.
Average time to first token. Lower is better for responsiveness.
| Rank | Provider | Model | First Token Latency | Avg tokens per second | Total Tests |
|---|---|---|---|---|---|
| 1 | glm-4-flash-250414 | 0.27 s Best: 0.21Worst: 0.38 | 49.81t/s | 5 | |
| 2 | glm-4-flashx | 0.29 s Best: 0.27Worst: 0.30 | 71.64t/s | 5 |
| 3 | O One APIgpt.fitue.cc | glm-4-air | 0.33 s Best: 0.30Worst: 0.38 | 65.96t/s | 5 |
| 4 | 智谱AI开放平台open.bigmodel.cn | glm-4-flash | 0.43 s Best: 0.34Worst: 1.10 | 40.94t/s | 10 |
| 5 |  Veloeralinjinpeng-veloera.hf.space | llama3.1-8b | 0.45 s Best: 0.40Worst: 0.52 | 1096.01t/s | 10 |
| 6 | Veloeralinjinpeng-veloera.hf.space | qwen-3-32b | 0.47 s Best: 0.45Worst: 0.51 | 733.22t/s | 5 |
| 7 | 3 30267340.cn-shanghai.pai-eas.aliyuncs.com30267340.cn-shanghai.pai-eas.aliyuncs.com | DeepSeek-R1-0528 | 0.47 s Best: 0.45Worst: 0.50 | 48.24t/s | 5 |
| 8 | Veloeralinjinpeng-veloera.hf.space | llama-4-109b | 0.48 s Best: 0.44Worst: 0.58 | 969.99t/s | 15 |
| 9 | Veloeralinjinpeng-veloera.hf.space | llama-3.3-70b | 0.52 s Best: 0.46Worst: 0.70 | 970.00t/s | 10 |
| 10 | O One APIgpt.fitue.cc | glm-4-flash | 0.55 s Best: 0.40Worst: 0.77 | 39.25t/s | 5 |
| 11 | J Joyueai.joyue.joyuerpa.com:3001 | deepseek-r1:32b | 0.56 s Best: 0.47Worst: 0.84 | 42.91t/s | 5 |
| 12 | a ai.joyue.joyuerpa.com:3001ai.joyue.joyuerpa.com:3001 | deepseek-r1:32b | 0.56 s Best: 0.47Worst: 0.84 | 42.91t/s | 5 |
| 13 |  SiliconFlowapi.siliconflow.cn | Pro/THUDM/glm-4-9b-chat | 0.56 s Best: 0.51Worst: 0.61 | 75.87t/s | 5 |
| 14 | SiliconFlowapi.siliconflow.cn | Pro/Qwen/Qwen2-7B-Instruct | 0.56 s Best: 0.52Worst: 0.61 | 71.29t/s | 5 |
| 15 | a api.openai-sb.comapi.openai-sb.com | gpt-4o-mini | 0.57 s Best: 0.49Worst: 0.65 | 63.02t/s | 5 |
| 16 | N Neo APIapi.chiban.de | DeepSeek-V3-Fast | 0.60 s Best: 0.54Worst: 0.62 | 102.13t/s | 5 |
| 17 | d d06200723-qwen330b-a3b-2160-b8lglles-11434.550c.cloudd06200723-qwen330b-a3b-2160-b8lglles-11434.550c.cloud | qwen3:30b-a3b | 0.60 s Best: 0.43Worst: 1.01 | 44.22t/s | 5 |
| 18 |  共绩算力d06200723-qwen330b-a3b-2160-b8lglles-11434.550c.cloud | qwen3:30b-a3b | 0.60 s Best: 0.43Worst: 1.01 | 44.22t/s | 5 |
| 19 | SiliconFlowapi.siliconflow.cn | Qwen/Qwen2-7B-Instruct | 0.61 s Best: 0.59Worst: 0.62 | 85.82t/s | 5 |
| 20 |  Novita AIapi.novita.ai | deepseek/deepseek-r1-0528 | 0.71 s Best: 0.52Worst: 1.17 | 60.77t/s | 10 |
| 21 | SiliconFlowapi.siliconflow.cn | Qwen/Qwen2.5-7B-Instruct | 0.71 s Best: 0.62Worst: 0.80 | 16.79t/s | 5 |
| 22 | 1 154.17.230.220:3000154.17.230.220:3000 | gpt-4.1-mini | 0.73 s Best: 0.56Worst: 1.21 | 156.81t/s | 50 |
| 23 | r realpics.cn:2234realpics.cn:2234 | qwen3-30b-a3b | 0.74 s Best: 0.69Worst: 0.77 | 62.84t/s | 5 |
| 24 | R Realpicsrealpics.cn:2234 | qwen3-30b-a3b | 0.74 s Best: 0.69Worst: 0.77 | 62.84t/s | 5 |
| 25 | r realpics.cn:2234realpics.cn:2234 | qwen3-30b-a3b | 0.74 s Best: 0.69Worst: 0.77 | 62.84t/s | 5 |
| 26 | A AI Toolsplatform.aitools.cfd | zhipu/glm-4v-flash | 0.74 s Best: 0.46Worst: 1.03 | 58.80t/s | 5 |
| 27 | O One APIgpt.fitue.cc | gemini-2.0-flash | 0.74 s Best: 0.68Worst: 0.84 | 149.12t/s | 5 |
| 28 |  DashScopedashscope.aliyuncs.com | qwen2.5-coder-0.5b-instruct | 0.75 s Best: 0.71Worst: 0.82 | 60.38t/s | 5 |
| 29 | O One APIgpt.fitue.cc | yi-lightning | 0.79 s Best: 0.66Worst: 1.28 | 43.78t/s | 5 |
| 30 | 1 154.17.230.220:3000154.17.230.220:3000 | gpt-4.1 | 0.80 s Best: 0.61Worst: 1.02 | 154.62t/s | 25 |
| 31 | 黑 黑名单拦截oneapi.k4xz7d6xxp.orange233.top | qwen3:0.6b | 0.89 s Best: 0.48Worst: 2.98 | 63.46t/s | 10 |
| 32 | DashScopedashscope.aliyuncs.com | qwen-coder-plus-latest | 0.96 s Best: 0.67Worst: 1.96 | 53.02t/s | 5 |
| 33 | DashScopedashscope.aliyuncs.com | qwen-turbo-latest | 0.96 s Best: 0.63Worst: 1.94 | 49.84t/s | 5 |
| 34 | DashScopedashscope.aliyuncs.com | qwen2.5-72b-instruct | 0.97 s Best: 0.73Worst: 1.85 | 28.17t/s | 5 |
| 35 | A AI Toolsplatform.aitools.cfd | zhipu/glm-4-flash | 0.99 s Best: 0.40Worst: 3.53 | 37.51t/s | 115 |
| 36 | 国 国信新网zygf.guoxincloud.cn:1025 | deepseekv3 | 1.00 s Best: 0.63Worst: 2.97 | 8.54t/s | 30 |
| 37 | 国 国信新网zygf.guoxincloud.cn:1025 | deepseekv3 | 1.00 s Best: 0.63Worst: 2.97 | 8.54t/s | 30 |
| 38 | DashScopedashscope.aliyuncs.com | qwen2.5-coder-3b-instruct | 1.00 s Best: 0.72Worst: 1.89 | 38.45t/s | 5 |
| 39 | d d06131241-vllmbasev03-201-v5gshhys-8000.550c.cloudd06131241-vllmbasev03-201-v5gshhys-8000.550c.cloud | deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B | 1.00 s Best: 0.32Worst: 3.66 | 194.01t/s | 5 |
| 40 | 共绩算力d06131241-vllmbasev03-201-v5gshhys-8000.550c.cloud | deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B | 1.00 s Best: 0.32Worst: 3.66 | 194.01t/s | 5 |
| 41 | d d06131241-vllmbasev03-201-v5gshhys-8000.550c.cloudd06131241-vllmbasev03-201-v5gshhys-8000.550c.cloud | deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B | 1.00 s Best: 0.32Worst: 3.66 | 194.01t/s | 5 |
| 42 |  New APIapi.lianwusuoai.top | 沉浸式翻译 | 1.01 s Best: 0.79Worst: 1.27 | 72.24t/s | 5 |
| 43 | V Veloeraveloera.yixya.top | qwen-turbo-latest | 1.02 s Best: 0.66Worst: 1.92 | 39.71t/s | 10 |
| 44 |  V-APIcf.v36.cm | gpt-4.1-nano | 1.05 s Best: 0.90Worst: 1.22 | 152.73t/s | 5 |
| 45 |  V-APIapi.vveai.com | gpt-4.1-nano | 1.05 s Best: 0.81Worst: 1.57 | 137.68t/s | 10 |
| 46 | O One APIgpt.fitue.cc | hunyuan-turbos-latest | 1.29 s Best: 1.03Worst: 1.58 | 36.99t/s | 5 |
| 47 | V-APIapi.vveai.com | gpt-4o-mini | 1.29 s Best: 0.59Worst: 2.00 | 77.53t/s | 5 |
| 48 | 国 国信新网zygf.guoxincloud.cn:1025 | deepseekv3 | 1.34 s Best: 0.73Worst: 1.97 | 8.32t/s | 5 |
| 49 | 国 国信新网zygf.guoxincloud.cn:1025 | deepseekv3 | 1.34 s Best: 0.73Worst: 1.97 | 8.32t/s | 5 |
| 50 |  AAAIapi.aaai.vip | gemini-2.5-flash-lite-preview-06-17 | 1.38 s Best: 1.20Worst: 1.85 | 310.35t/s | 5 |