Leaderboard
Multi-dimensional rankings based on model speed tests and provider health checks. Compare providers, endpoints, and reliability at a glance.
Average tokens generated per second. Higher is better for fast responses.
| Rank | Provider | Model | Throughput | Avg first token latency | Total Tests |
|---|---|---|---|---|---|
| 1 | o3-mini-high | 141126.77 t/s Best: 190384.25Worst: 76217.99 | 79.37s | 5 | |
| 2 | o3-mini-2025-01-31 | 46653.59 t/s Best: 91395.50Worst: 13978.82 | 14.95s | 10 | |
| 3 |
ai.api.xn--fiqs8sai.api.xn--fiqs8s |
| BLOOMZ-7B |
875.69 t/s Best: 3707.84Worst: 115.27 |
2.46s |
| 5 |
| 4 | n new.wei.binew.wei.bi | 0311codelalama:latest | 624.78 t/s Best: 688.03Worst: 516.51 | 1.58s | 5 |
| 5 | ai.api.xn--fiqs8sai.api.xn--fiqs8s | gemini-1.5-flash-8b | 275.08 t/s Best: 392.08Worst: 168.24 | 1.28s | 10 |
| 6 | n new.wei.binew.wei.bi | deepseek-ai/DeepSeek-R1 | 261.08 t/s Best: 394.60Worst: 211.83 | 0.62s | 20 |
| 7 | ai.api.xn--fiqs8sai.api.xn--fiqs8s | gemini-1.5-flash-latest | 252.88 t/s Best: 961.70Worst: 152.64 | 1.26s | 10 |
| 8 | ai.api.xn--fiqs8sai.api.xn--fiqs8s | gemini-1.5-flash-002 | 239.65 t/s Best: 491.56Worst: 163.29 | 1.62s | 5 |
| 9 | ai.api.xn--fiqs8sai.api.xn--fiqs8s | gemini-2.0-flash-thinking-exp-01-21 | 230.82 t/s Best: 283.17Worst: 195.73 | 7.47s | 5 |
| 10 | ZetaTechs APIapi.zetatechs.com | gemini-2.0-flash-thinking-exp | 224.26 t/s Best: 254.06Worst: 188.25 | 7.53s | 5 |
| 11 | n new.wei.binew.wei.bi | deepseek-r1:1.5b | 217.41 t/s Best: 226.59Worst: 210.08 | 0.52s | 5 |
| 12 | ai.api.xn--fiqs8sai.api.xn--fiqs8s | o3-mini | 212.69 t/s Best: 449.71Worst: 21.31 | 7.80s | 10 |
| 13 | FastRouterapi.055ai.cn | gemini-ai/gemini-2.0-flash-lite-preview-02-05 | 211.57 t/s Best: 432.33Worst: 157.24 | 1.63s | 20 |
| 14 | MN APIwww.mnapi.com | gemini-2.0-flash-thinking-exp-01-21 | 194.98 t/s Best: 417.60Worst: 125.78 | 2.12s | 5 |
| 15 | N New APIapi.hongshi.me | gpt-4o-mini | 192.15 t/s Best: 511.52Worst: 5.12 | 8.99s | 5 |
| 16 | ai.api.xn--fiqs8sai.api.xn--fiqs8s | gemini-2.0-flash-lite-preview-02-05 | 188.38 t/s Best: 222.97Worst: 150.48 | 1.06s | 15 |
| 17 | V-APIapi.gpt.ge | gemini-1.5-flash-latest | 169.06 t/s Best: 203.29Worst: 154.83 | 1.02s | 5 |
| 18 | ai.api.xn--fiqs8sai.api.xn--fiqs8s | gemini-1.5-flash | 158.26 t/s Best: 171.97Worst: 147.19 | 0.91s | 5 |
| 19 | Dream APIopus.gptuu.com | gpt-3.5-turbo-1106 | 156.38 t/s Best: 159.97Worst: 147.56 | 0.82s | 5 |
| 20 | n nginxkfcv50.link | gpt-4o | 147.84 t/s Best: 284.02Worst: 64.05 | 9.22s | 10 |
| 21 | Dream APIopus.gptuu.com | gpt-3.5-turbo-16k | 145.74 t/s Best: 168.81Worst: 80.91 | 0.64s | 5 |
| 22 | Dream APIopus.gptuu.com | gpt-3.5-turbo-0613 | 140.29 t/s Best: 146.05Worst: 129.76 | 0.50s | 5 |
| 23 | ai.api.xn--fiqs8sai.api.xn--fiqs8s | gpt-4o | 135.41 t/s Best: 193.68Worst: 117.01 | 1.43s | 5 |
| 24 | Zeaburneapi.zeabur.app | deepseek-r1:7b | 132.96 t/s Best: 152.91Worst: 117.88 | 13.15s | 5 |
| 25 | Dream APIopus.gptuu.com | gpt-4o-mini-2024-07-18 | 130.72 t/s Best: 153.13Worst: 91.96 | 0.60s | 5 |
| 26 | ZEN-AI VIPvip.zen-ai.top | gpt-4o | 129.13 t/s Best: 801.45Worst: 30.42 | 5.78s | 10 |
| 27 | Zeaburneapi.zeabur.app | best-free | 128.14 t/s Best: 134.92Worst: 103.80 | 2.76s | 5 |
| 28 | DuckDuck APIduck2api.com | gpt-4o-2024-11-20 | 127.81 t/s Best: 149.68Worst: 86.04 | 0.91s | 5 |
| 29 | ai.api.xn--fiqs8sai.api.xn--fiqs8s | gemini-2.0-flash | 125.24 t/s Best: 139.25Worst: 103.13 | 1.07s | 5 |
| 30 | ai.api.xn--fiqs8sai.api.xn--fiqs8s | gemini-2.0-flash-exp | 125.02 t/s Best: 153.41Worst: 103.68 | 1.40s | 5 |
| 31 | FastRouterapi.055ai.cn | gemini-ai/gemini-2.0-flash | 124.50 t/s Best: 166.28Worst: 94.73 | 1.88s | 50 |
| 32 | Dream APIopus.gptuu.com | gpt-4o-2024-05-13 | 122.74 t/s Best: 144.33Worst: 102.60 | 1.60s | 5 |
| 33 | b bljj.orgbljj.org | gemini-2.0-flash | 122.53 t/s Best: 156.09Worst: 95.92 | 1.44s | 10 |
| 34 | MN APIwww.mnapi.com | gemini-2.0-flash | 121.90 t/s Best: 164.19Worst: 80.06 | 2.23s | 10 |
| 35 | AIO通用智能服务平台api-s1.aiearth.vip | gemini-2.0-flash-thinking-exp-01-21 | 121.89 t/s Best: 138.95Worst: 109.19 | 1.38s | 5 |
| 36 | 毫秒APIapi.holdai.top | gemini-2.0-flash-exp | 121.75 t/s Best: 142.66Worst: 106.18 | 10.83s | 5 |
| 37 | Dream APIopus.gptuu.com | gpt-3.5-turbo | 121.66 t/s Best: 134.38Worst: 103.80 | 0.79s | 5 |
| 38 | 共绩算力550c.cloud | deepseek-r1:7b | 119.65 t/s Best: 141.53Worst: 5.79 | 9.09s | 120 |
| 39 | C C Z0 Api 01c-z0-api-01.hash070.com | gemini-2.0-flash | 113.79 t/s Best: 129.94Worst: 100.58 | 0.90s | 5 |
| 40 | 算 算了么 APIapi.suanli.cn | deepseek-r1:7b | 109.24 t/s Best: 132.73Worst: 76.68 | 15.14s | 5 |
| 41 | o ocool AIapi.ocoolai.com | deepseek-ai/DeepSeek-R1 | 108.28 t/s Best: 324.23Worst: 20.13 | 2.67s | 15 |
| 42 | Dream APIopus.gptuu.com | gpt-3.5-turbo-16k-0613 | 108.01 t/s Best: 160.63Worst: 82.92 | 0.73s | 5 |
| 43 | Dream APIopus.gptuu.com | gpt-4o-2024-08-06 | 107.84 t/s Best: 117.96Worst: 99.92 | 3.23s | 5 |
| 44 | ai.api.xn--fiqs8sai.api.xn--fiqs8s | qwen-72b | 102.22 t/s Best: 108.02Worst: 98.38 | 3.11s | 5 |
| 45 | Dream APIopus.gptuu.com | gpt-4o | 101.73 t/s Best: 160.17Worst: 80.30 | 1.24s | 10 |
| 46 | Dream APIopus.gptuu.com | gpt-4o-mini | 101.71 t/s Best: 143.55Worst: 43.03 | 0.68s | 5 |
| 47 | MN APIwww.mnapi.com | gpt-4o-2024-08-06 | 99.32 t/s Best: 135.15Worst: 55.08 | 2.37s | 10 |
| 48 | Dream APIopus.gptuu.com | gpt-3.5-turbo-0125 | 98.60 t/s Best: 142.27Worst: 0.00 | 0.70s | 5 |
| 49 | G Gpt APIgpt-api.cc | gpt-4o-mini | 94.27 t/s Best: 112.37Worst: 84.11 | 1.92s | 5 |
| 50 | ePhone AIapi.ephone.ai | gpt-4o-mini | 93.51 t/s Best: 115.74Worst: 76.91 | 1.08s | 5 |