Ollama Cloud tokens per second — live benchmark
Real inference speed, measured continuously. Every row is a live Ollama Cloud model — sorted by tokens per second, benchmarked every ~10 minutes.
Weighing Ollama Cloud against a router? Ollama Cloud vs OpenRouter — what we measure, and what we don't.
| Trend | |||||||
|---|---|---|---|---|---|---|---|
| Muse Spark Open Pro | 336.5 | 160.7 | — | 100% | 24.9 | | 55s ago |
| GPT-OSS 120B Pro | 306.6 | 388.3 | 349ms | 100% | 3.8 | | 8m ago |
| Gemma4 31B Pro | 270.8 | 261.3 | 320ms | 100% | 4.4 | | 5m ago |
| MiMo V2.6 Pro Pro | 240.2 | 220.9 | 396ms | 100% | 29.9 | | 2m ago |
| MiMo V2.6 Flash Pro | 236.9 | 277.3 | 353ms | 100% | 26.7 | | 4m ago |
| DeepSeek V4.1 Flash Pro | 213.2 | 259.3 | 483ms | 100% | — | | 10m ago |
| DeepSeek V4.1 Flash 1017 Pro | 208.3 | 242.0 | 536ms | 100% | 32.6 | | 8m ago |
| GLM 5.3 Flash Pro | 188.8 | 211.3 | 511ms | 100% | 27.8 | | 10m ago |
| GLM 5.3 Pro | 183.2 | 225.3 | 339ms | 100% | 31.7 | | 8m ago |
| GLM 5.2 Pro | 177.5 | 172.6 | 562ms | 100% | — | | 6m ago |
| MiniMax M3.1 Flash Pro | 170.7 | 246.2 | 508ms | 100% | 29.1 | | 7m ago |
| Nemotron 3 Super Pro | 152.3 | 119.0 | 402ms | 99% | — | | 7m ago |
| Beam Pro | 137.9 | 161.3 | 816ms | 100% | 17.7 | | 5m ago |
| Kimi K2.7 Code Pro | 134.0 | 163.8 | 673ms | 100% | — | | 2m ago |
| DeepSeek V4 Pro 0813 Pro | 131.5 | 133.6 | 604ms | 100% | 23.1 | | 5m ago |
| GPT-OSS 20B Pro | 129.5 | 154.2 | 372ms | 100% | — | | 5m ago |
| Kimi K2.6 Pro | 126.0 | 115.9 | 511ms | 100% | — | | 2m ago |
| MiniMax M3 Pro | 112.2 | 155.6 | 816ms | 100% | 18.9 | | 7m ago |
| Mistral Large 4 Pro | 109.5 | 88.6 | 2.0s | 99% | 26.1 | | 5m ago |
| Kimi K3 Pro | 87.3 | 85.5 | 944ms | 100% n=6 | — | | 3h ago |
| Kimi K3.1 Pro Pro | 87.1 | 65.5 | 839ms | 99% | 46.5 | | 5m ago |
| Nemotron 3 Ultra Pro | 32.6 | 10.9 | 735ms | 97% | 11.8 | | 1m ago |
No models match your filter.
Intelligence Index scores from Artificial Analysis.
0 models unavailable or stale
Support the live benchmarks
Frequently asked questions
What is the fastest Ollama Cloud model right now?
As of the last build, GPT-OSS 120B is the fastest Ollama Cloud model on this leaderboard at 388.3 tokens/sec (most recent measured run). Rankings change as models are re-benchmarked roughly every 10 minutes — see the live leaderboard above for the current order.
How is tokens per second measured?
TPS is generation throughput: output tokens divided by server-reported generation time, excluding time-to-first-token. For Ollama Cloud we use the server’s own reported timing (eval_count and total_duration) rather than a client-side stopwatch, so results are immune to network jitter and token buffering. Full formula and error handling are on the methodology page.
How often is data refreshed?
The worker benchmarks continuously using a round-robin priority queue. Ollama Cloud Pro is sampled about every 10 minutes. Kimi K3 retains a conservative target of about every 4 hours. The leaderboard also polls the API roughly every 60 seconds client-side to keep cells current without a full page reload.