Ollama Cloud models
Live speed benchmarks for every model available on Ollama Cloud. Numbers update every ~10 minutes (Ollama Free hourly) — sorted by latest tokens per second.
All Ollama Cloud models by speed
| Model | TPS now | TPS 24h avg | TTFT | Reliability | Intelligence Index |
|---|---|---|---|---|---|
| 1 GPT-OSS 120B | 362.6 | 333.9 | 338ms | 100% | 11.6 |
| 2 Gemma4 31B | 291.4 | 235.3 | 291ms | 100% | 14.7 |
| 3 GPT-OSS 120B | 290.8 | 290.9 | 362ms | 100% | 11.6 |
| 4 Gemma4 31B | 289.1 | 220.1 | 325ms | 100% | 14.7 |
| 5 GLM 5.3 Flash | 266.6 | 217.8 | 497ms | 100% | 41.8 |
| 6 MiMo V2.6 Pro | 263.9 | 236.6 | 697ms | 100% | 46.3 |
| 7 MiniMax M3 | 223.6 | 120.7 | 755ms | 100% | 29.2 |
| 8 DeepSeek V4.1 Flash 1017 | 216.7 | 211.4 | 584ms | 100% | 45.1 |
| 9 MiMo V2.6 Flash | 213.1 | 256.7 | 309ms | 100% | 37.9 |
| 10 Nemotron 3 Super | 212.9 | 126.3 | 378ms | 100% | 12.8 |
| 11 GLM 5.2 | 201.7 | 185.0 | 382ms | 100% | 33.7 |
| 12 DeepSeek V4.1 Flash | 200.1 | 221.5 | 459ms | 100% | 39.5 |
| 13 Muse Spark Open | 190.4 | 195.3 | — | 100% | 38.9 |
| 14 Kimi K2.6 | 189.6 | 167.8 | 439ms | 100% | 27 |
| 15 DeepSeek V4 Pro 0813 | 174.3 | 151.4 | 454ms | 100% | 36 |
| 16 Nemotron 3 Nano 30B (non-reasoning) | 167.2 | 227.3 | 256ms | 100% | 6.8 |
| 17 Nemotron 3 Nano 30B (non-reasoning) | 140.6 | 230.1 | 1.1s | 100% | 6.8 |
| 18 GPT-OSS 20B | 135.3 | 119.8 | 369ms | 100% | 9 |
| 19 GPT-OSS 20B | 128.8 | 133.8 | 376ms | 100% | 9 |
| 20 Kimi K2.7 Code | 128.7 | 134.0 | 706ms | 100% | 25.8 |
| 21 Nemotron 3 Super | 120.1 | 129.8 | 459ms | 100% | 12.8 |
| 22 Beam | 117.0 | 105.3 | 504ms | 100% | 30.3 |
| 23 Kimi K3 | 107.9 | 110.0 | 570ms | 100% | 43.6 |
| 24 Mistral Large 3 675B (non-reasoning) | 97.3 | 86.9 | 2.2s | 100% | 9.3 |
| 25 Beam | 95.3 | 100.4 | 421ms | 100% | 30.3 |
| 26 Nemotron 3 Ultra | 79.4 | 43.2 | 716ms | 100% | 22.9 |
| 27 GLM 5.3 | 79.3 | 161.2 | 441ms | 100% | 44.8 |
| 28 MiniMax M3.1 Flash | 78.0 | 155.9 | 582ms | 100% | 42.7 |
| 29 Kimi K3.1 Pro | 67.1 | 67.7 | 1.9s | 100% | — |
| 30 Nemotron 3 Ultra | 56.9 | 47.8 | 1.0s | 100% | 22.9 |
| 31 MiniMax M2.7 | 54.0 | 52.6 | 895ms | 100% | 22.8 |
| 32 Mistral Large 4 | 28.5 | 96.9 | 7.0s | 99% | 38.4 |
About Ollama Cloud models
Ollama Cloud provides access to large language models via a hosted API. Models range from lightweight coding assistants (e.g. gpt-oss:20b) to flagship reasoning models (e.g. deepseek-v4-pro), each with its own per-million-token rate on the pricing page. Local model usage is always unlimited; cloud models spend your plan's monthly usage credits.
The benchmarks on this page come from continuous automated tests — one streaming chat completion per model per ~10 minutes, measured from outside Ollama's network. See the methodology page for the full measurement spec.
Compare models side by side
Use the compare tool to overlay speed timelines for up to 6 models. Useful for picking between similar-sized models or tracking performance changes over time.