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 | gpt-5-nano | 52788.85 t/s Best: 114460.71Worst: 12345.63 | 26.15s | 5 | |
| 2 | deepseek-ai/DeepSeek-V3.1 | 37419.93 t/s Best: 187248.67Worst: 36.95 | 3.04s | 15 | |
| 3 |
DashScopedashscope.aliyuncs.com |
| qwen-mt-turbo |
5403.35 t/s Best: 8067.50Worst: 3435.35 |
1.02s |
| 5 |
| 4 | G GPT Loadallaiload.dpdns.org | qwen-3-32b | 3044.37 t/s Best: 3805.59Worst: 2429.74 | 1.93s | 5 |
| 5 | G GPT Loadallaiload.dpdns.org | qwen-3-235b-a22b-instruct-2507 | 2212.66 t/s Best: 5983.65Worst: 1432.88 | 1.99s | 10 |
| 6 | S SkyAIskyai.089apis.xyz | gpt-oss-120b | 671.95 t/s Best: 814.13Worst: 569.76 | 2.33s | 5 |
| 7 | RinkoAIrinkoai.com | gpt-oss-120b | 590.03 t/s Best: 822.90Worst: 458.71 | 0.84s | 5 |
| 8 | YUNWU APIyunwu.ai | mistral-small-latest | 576.27 t/s Best: 807.33Worst: 70.75 | 1.51s | 5 |
| 9 | New APIfanyi.963312.xyz | qwen-3-235b-a22b-instruct-2507 | 449.95 t/s Best: 936.84Worst: 109.20 | 1.47s | 5 |
| 10 | Gemini Balancehmcldastbscm.ap-northeast-1.clawcloudrun.com | gemini-flash-lite-latest | 381.81 t/s Best: 517.92Worst: 290.49 | 0.88s | 20 |
| 11 | G GPT Loadallaiload.dpdns.org | qwen-3-235b-a22b-instruct-2507 | 377.69 t/s Best: 420.23Worst: 282.14 | 1.38s | 5 |
| 12 | SkyAIapi.071572.xyz | gpt-oss-120b | 365.50 t/s Best: 827.27Worst: 194.05 | 2.12s | 5 |
| 13 | G GPT Loadallaiload.dpdns.org | translate-model | 345.18 t/s Best: 2297.88Worst: 48.59 | 3.10s | 45 |
| 14 | YUNWU APIyunwu.ai | gemini-2.5-flash-lite-nothinking | 341.00 t/s Best: 397.96Worst: 261.97 | 0.99s | 5 |
| 15 | New APIfanyi.963312.xyz | gpt-oss-120b | 306.96 t/s Best: 776.89Worst: 95.72 | 1.64s | 5 |
| 16 | YUNWU APIyunwu.zeabur.app | gemini-2.5-flash-lite-nothinking | 295.30 t/s Best: 391.82Worst: 236.11 | 0.99s | 5 |
| 17 | G GPT Loadallaiload.dpdns.org | models/gemini-2.5-flash-lite | 259.93 t/s Best: 299.06Worst: 233.31 | 0.71s | 5 |
| 18 | DashScopedashscope.aliyuncs.com | qwen3-0.6b | 243.65 t/s Best: 289.74Worst: 204.64 | 2.38s | 5 |
| 19 | S SkyAIskyai.089apis.xyz | moonshotai/kimi-k2-instruct-0905 | 240.47 t/s Best: 294.02Worst: 168.89 | 1.79s | 10 |
| 20 | YUNWU APIyunwu.ai | gpt-5-nano | 238.85 t/s Best: 376.48Worst: 134.58 | 7.16s | 10 |
| 21 | RinkoAIrinkoai.com | moonshotai/kimi-k2-instruct-0905 | 227.97 t/s Best: 297.72Worst: 182.46 | 0.78s | 5 |
| 22 | K K2Thinkk2t.shiho.top | MBZUAI-IFM/K2-Think-nothink | 219.47 t/s Best: 234.00Worst: 196.83 | 2.56s | 5 |
| 23 | G GPT Loadallaiload.dpdns.org | DeepSeek-V3-0324 | 218.28 t/s Best: 265.24Worst: 119.32 | 0.90s | 5 |
| 24 | 1 180.76.61.29:3000180.76.61.29:3000 | Qwen/Qwen3-Next-80B-A3B-Instruct | 215.77 t/s Best: 259.45Worst: 179.04 | 1.75s | 5 |
| 25 | New APInew.123nhh.xyz | gemini-2.5-flash | 210.68 t/s Best: 243.31Worst: 183.09 | 7.27s | 5 |
| 26 | N New APIwxkyw.dpdns.org | gemini-2.5-flash | 206.02 t/s Best: 249.82Worst: 162.88 | 7.86s | 5 |
| 27 | G GPT Loadallaiload.dpdns.org | openai/gpt-oss-120b | 205.99 t/s Best: 224.04Worst: 194.48 | 8.90s | 5 |
| 28 | G GPT Loadallaiload.dpdns.org | models/gemini-2.5-flash | 198.97 t/s Best: 254.68Worst: 161.26 | 7.95s | 15 |
| 29 | EU.orgcpapi.apkipa.eu.org | gemini-2.5-flash | 191.67 t/s Best: 228.77Worst: 148.61 | 10.04s | 5 |
| 30 | EU.orgcpapi.apkipa.eu.org | gemini-2.5-flash | 191.67 t/s Best: 228.77Worst: 148.61 | 10.04s | 5 |
| 31 | ModelScopeapi-inference.modelscope.cn | Qwen/Qwen3-Next-80B-A3B-Instruct | 178.62 t/s Best: 237.70Worst: 138.07 | 0.97s | 5 |
| 32 | DashScopedashscope.aliyuncs.com | qwen3-1.7b | 175.64 t/s Best: 192.92Worst: 152.76 | 4.38s | 5 |
| 33 | G GPT Loadallaiload.dpdns.org | Qwen/Qwen3-Next-80B-A3B-Instruct | 173.09 t/s Best: 227.84Worst: 139.15 | 1.54s | 5 |
| 34 | ChatGTPwww.chatgtp.cn | gemini-2.0-flash | 170.98 t/s Best: 254.40Worst: 119.27 | 1.37s | 5 |
| 35 | 智谱AI开放平台open.bigmodel.cn | glm-z1-airx | 168.24 t/s Best: 209.25Worst: 147.32 | 0.34s | 5 |
| 36 | New API20230621.xyz | gemini-2.5-flash | 165.42 t/s Best: 210.27Worst: 123.48 | 0.86s | 5 |
| 37 | Veloeralinjinpeng-veloera.hf.space | momentum | 163.91 t/s Best: 213.50Worst: 87.86 | 6.20s | 5 |
| 38 | A AI Toolsplatform.aitools.cfd | openai/gpt-oss-20b | 154.97 t/s Best: 353.08Worst: 0.00 | 2.79s | 20 |
| 39 | G GPT Loadallaiload.dpdns.org | qwen/qwen3-next-80b-a3b-instruct | 154.28 t/s Best: 168.44Worst: 130.24 | 0.79s | 10 |
| 40 | integrate.api.nvidia.comintegrate.api.nvidia.com | openai/gpt-oss-120b | 149.96 t/s Best: 171.49Worst: 130.58 | 18.38s | 5 |
| 41 | TokenPonyapi.tokenpony.cn | hunyuan-a13b-instruct | 143.03 t/s Best: 143.55Worst: 141.86 | 3.85s | 5 |
| 42 | G GPT Loadallaiload.dpdns.org | gpt-oss:120b | 141.34 t/s Best: 164.35Worst: 119.92 | 1.35s | 5 |
| 43 | G GPT Loadallaiload.dpdns.org | WiNGPT-Babel | 137.86 t/s Best: 288.46Worst: 71.33 | 1.40s | 5 |
| 44 | New APInewapi.df-h.com | immersive_translate | 127.41 t/s Best: 137.07Worst: 112.45 | 0.31s | 10 |
| 45 | 线衣apixianyi.zeabur.app | openai/gpt-oss-120b | 127.08 t/s Best: 152.09Worst: 79.75 | 1.81s | 5 |
| 46 | 腾讯云api.hunyuan.cloud.tencent.com | hunyuan-lite | 123.45 t/s Best: 131.43Worst: 114.42 | 0.88s | 5 |
| 47 | ModelScopeapi-inference.modelscope.cn | Qwen/Qwen3-30B-A3B | 123.38 t/s Best: 153.33Worst: 93.68 | 6.13s | 5 |
| 48 | C CharTyrapi.char.icu | doubao-seed-1.6-flash | 112.37 t/s Best: 149.80Worst: 83.62 | 7.99s | 5 |
| 49 | O OpenRestyapi.longcat.chat | LongCat-Flash-Chat | 110.44 t/s Best: 165.84Worst: 82.89 | 2.94s | 10 |
| 50 | 1 180.76.61.29:3000180.76.61.29:3000 | Qwen/Qwen3-Next-80B-A3B-Instruct | 110.28 t/s Best: 147.55Worst: 83.24 | 0.95s | 5 |