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-5.2 | 606.84 t/s Best: 900.16Worst: 294.43 | 11.49s | 10 | |
| 2 | gpt-5.2 | 604.11 t/s Best: 904.25Worst: 230.58 | 12.57s | 20 | |
| 3 |
| test |
583.14 t/s Best: 613.12Worst: 524.38 |
0.46s |
| 5 |
| 4 | zai-glm-4.7 | 454.25 t/s Best: 609.14Worst: 295.71 | 3.57s | 5 |
| 5 | meta-llama/Llama-3.3-70B-Instruct | 416.19 t/s Best: 538.29Worst: 336.89 | 0.26s | 5 |
| 6 | meta-llama/Llama-3.3-70B-Instruct | 416.19 t/s Best: 538.29Worst: 336.89 | 0.26s | 5 |
| 7 | gpt-oss-120b-medium | 337.69 t/s Best: 393.18Worst: 266.67 | 2.32s | 20 |
| 8 | ministral-3b-2410 | 332.18 t/s Best: 539.88Worst: 217.45 | 0.50s | 5 |
| 9 | ministral-3b-2410 | 332.18 t/s Best: 539.88Worst: 217.45 | 0.50s | 5 |
| 10 | gpt-5-codex-mini | 327.07 t/s Best: 407.63Worst: 157.66 | 3.35s | 5 |
| 11 | gemini-2.5-flash-lite | 298.48 t/s Best: 370.05Worst: 234.32 | 1.75s | 5 |
| 12 | gemini-2.5-flash-lite | 297.64 t/s Best: 345.16Worst: 255.19 | 0.52s | 5 |
| 13 | gemini-2.5-flash-lite | 269.50 t/s Best: 381.10Worst: 139.05 | 1.01s | 10 |
| 14 | gpt-5.1-codex-mini | 248.33 t/s Best: 265.97Worst: 214.97 | 1.91s | 5 |
| 15 | gemini-2.5-flash-lite | 234.75 t/s Best: 359.28Worst: 114.50 | 2.52s | 30 |
| 16 | open-mistral-nemo | 200.44 t/s Best: 224.63Worst: 168.65 | 0.40s | 5 |
| 17 | open-mistral-nemo | 200.44 t/s Best: 224.63Worst: 168.65 | 0.40s | 5 |
| 18 | gemini-3-pro-preview-search | 199.11 t/s Best: 422.28Worst: 110.87 | 15.69s | 5 |
| 19 | gemini-3-flash-preview | 196.95 t/s Best: 626.69Worst: 126.65 | 9.59s | 10 |
| 20 | 酒馆-Flash-Long | 194.62 t/s Best: 212.42Worst: 179.41 | 1.75s | 5 |
| 21 | gemini-2.5-flash | 194.27 t/s Best: 260.28Worst: 147.23 | 9.13s | 5 |
| 22 | [官逆C]gemini-3-flash-preview | 193.84 t/s Best: 258.15Worst: 138.52 | 5.57s | 5 |
| 23 | magistral-small-latest | 189.91 t/s Best: 230.24Worst: 160.56 | 0.39s | 5 |
| 24 | magistral-small-latest | 189.91 t/s Best: 230.24Worst: 160.56 | 0.39s | 5 |
| 25 | gemini-2.5-flash | 189.12 t/s Best: 231.86Worst: 143.35 | 11.43s | 5 |
| 26 | accounts/fireworks/models/minimax-m2p1 | 185.81 t/s Best: 216.27Worst: 154.65 | 1.91s | 10 |
| 27 | gemini-3-flash-preview-search | 184.83 t/s Best: 276.49Worst: 117.50 | 9.96s | 5 |
| 28 | gpt-4o | 182.75 t/s Best: 245.77Worst: 143.71 | 3.58s | 5 |
| 29 | gemini-2.0-flash | 175.01 t/s Best: 192.51Worst: 154.00 | 0.56s | 5 |
| 30 | qwen3-1.7b | 171.33 t/s Best: 184.46Worst: 155.83 | 3.66s | 10 |
| 31 | gemini-2.0-flash | 171.32 t/s Best: 194.98Worst: 137.78 | 1.85s | 5 |
| 32 | gemini-2.0-flash | 171.32 t/s Best: 194.98Worst: 137.78 | 1.85s | 5 |
| 33 | gemini-3-flash-preview | 170.88 t/s Best: 221.88Worst: 135.19 | 6.15s | 15 |
| 34 | gemini-3-flash-preview | 164.68 t/s Best: 222.60Worst: 114.65 | 8.71s | 10 |
| 35 | gemini-3-flash-preview | 159.56 t/s Best: 217.53Worst: 124.51 | 6.93s | 10 |
| 36 | models/gemini-3-flash-preview | 156.97 t/s Best: 174.49Worst: 134.99 | 4.02s | 5 |
| 37 | gemini-3-flash-preview | 155.47 t/s Best: 176.06Worst: 142.99 | 7.06s | 5 |
| 38 | mistral-tiny-latest | 155.43 t/s Best: 216.68Worst: 110.17 | 0.38s | 5 |
| 39 | mistral-tiny-latest | 155.43 t/s Best: 216.68Worst: 110.17 | 0.38s | 5 |
| 40 | gemini-2.5-computer-use-preview-10-2025 | 154.96 t/s Best: 177.60Worst: 119.27 | 10.46s | 5 |
| 41 | gemini-3-flash-preview | 154.15 t/s Best: 170.82Worst: 131.68 | 10.44s | 5 |
| 42 | gemini-3-flash-preview | 144.88 t/s Best: 330.44Worst: 92.46 | 10.27s | 10 |
| 43 | accounts/fireworks/models/gpt-oss-120b | 144.71 t/s Best: 179.95Worst: 127.43 | 0.79s | 10 |
| 44 | o3-mini | 144.12 t/s Best: 163.66Worst: 95.05 | 4.81s | 5 |
| 45 | gemini-3-pro-preview | 143.36 t/s Best: 208.77Worst: 101.14 | 15.73s | 5 |
| 46 | qwen-flash | 142.89 t/s Best: 158.26Worst: 128.60 | 0.51s | 5 |
| 47 | gpt-5 | 141.76 t/s Best: 235.83Worst: 46.80 | 3.80s | 5 |
| 48 | claude-haiku-4-5-20251001 | 141.23 t/s Best: 177.66Worst: 89.72 | 1.92s | 5 |
| 49 | gemini-2.0-flash | 140.17 t/s Best: 168.04Worst: 119.68 | 0.86s | 15 |
| 50 | GLM-4.6V-Flash | 138.89 t/s Best: 227.32Worst: 80.80 | 10.30s | 10 |