Ollama Cloud
permanent free · inference_host · verification: partially_verified (70%, doc_fetch, 2026-07-17)
The record
| Base URLs | https://ollama.com/api native |
|---|---|
| API key required | yes |
| Credit card | unknown |
| Phone verification | unknown |
| May train on prompts | no |
| Quota | free_tier [concurrency, fair_use] “Ollama Cloud measures usage by GPU time, not tokens or requests. Free tier described as 'light usage' with session limits resetting every 5 hours and weekly limits every 7 days. Free plan: 1 cloud model at a time (concurrency).” |
| Signup friction | 3 steps · ~5m to first key |
| Env var | OLLAMA_API_KEY |
| Notes | Hosted service — distinct from local Ollama (which would be category 'local'). NOT OpenAI SDK-compatible: uses the native Ollama API (mnfst footnote 3; confirmed docs describe only the Ollama API at https://ollama.com/api). Note base URL discrepancy: mnfst lists https://api.ollama.com, official docs show https://ollama.com/api/ — official docs value recorded. ~36 cloud models available (Ollama library, cloud tag); no vision-capable cloud model confirmed in sources. |
Models (4)
| model | capabilities | context | free | limits |
|---|---|---|---|---|
| deepseek-v3.1:671b-cloud | chat | 131,072 | yes | Session/weekly limits (unpublished) |
| gpt-oss:120b-cloud | chat | 131,072 | yes | Session/weekly limits (unpublished) |
| kimi-k2:1t-cloud | chat | 268,288 | yes | Session/weekly limits (unpublished) |
| qwen3-coder:480b-cloud | chat | 131,072 | yes | Session/weekly limits (unpublished) |
Provenance
| source_ollama_cloud_docs | official_pricing | trust: high | checked 2026-07-17 |
| source_repo_mnfst | curated_repo | trust: medium_high | checked 2026-07-17 |