Ollama

"ollama cloud is disabled: remote model details are unavailable"

Last checked

Ollama is running with cloud access turned off (OLLAMA_NO_CLOUD=1, or disable_ollama_cloud set to true in ~/.ollama/server.json) while your config references a cloud or remote model, so its details can't be fetched. Either switch to a locally pulled model and remove the cloud ones with ollama rm, or, if you do want remote models, clear both settings, restart Ollama and sign in with ollama signin.

Why it happens

Ollama is running with cloud access turned off (OLLAMA_NO_CLOUD=1 or disable_ollama_cloud in ~/.ollama/server.json) while you referenced a cloud / remote model, so its details cannot be fetched.

  1. OLLAMA_NO_CLOUD=1 is set in the environment of the Ollama server, for example in a systemd unit, via launchctl on macOS, or as a Docker -e flag.
  2. ~/.ollama/server.json contains "disable_ollama_cloud": true. Turning off the Cloud switch in the Ollama app's Settings writes this file. Either source alone disables cloud.
  3. Your agent or client asked for the details of a cloud model, one whose name ends in :cloud or -cloud, through /api/show or /v1/models/<model>. Model discovery does this, and Ollama answers with HTTP 403 and this message.
  4. If the client tries to run the cloud model instead of looking it up, the same policy produces the variant "ollama cloud is disabled: remote model is unavailable".

The fix

  1. 1 Use a locally pulled model instead, or remove the cloud models (ollama rm <model>).
  2. 2 If you do want remote models, unset OLLAMA_NO_CLOUD (or set it to 0) and make sure ~/.ollama/server.json does not set disable_ollama_cloud to true (or turn Cloud back on in the app's Settings).
  3. 3 Restart Ollama, then sign in with ollama signin.
ollama pull qwen3-coder:30b

Still failing?

  • Restart Ollama and read its startup log. The line "Ollama cloud disabled: true" means the server still sees cloud as off.
  • Check both switches: the OLLAMA_NO_CLOUD value in the server's own environment (systemd unit, launchctl or container), not just your shell, and the contents of ~/.ollama/server.json.
  • If the app's Cloud setting says the OLLAMA_NO_CLOUD environment variable is forcing cloud off, remove the variable where it is set; the toggle cannot override it.
  • Search your agent config for model names ending in :cloud or -cloud and run ollama ls to see which installed entries are cloud models.
  • With cloud re-enabled, run ollama signin. Cloud models need an ollama.com account.

Related errors

Hit a different error?

Paste any agent error and get the cause and fix in seconds.

Open the decoder

Frequently asked questions

Do I actually need cloud models, or can I stay local?

For most agent work a locally pulled model is enough. Pull a tool-capable model like qwen3-coder:30b and point your config at it to avoid the cloud dependency entirely.

I do want the remote model. How do I re-enable cloud?

Unset OLLAMA_NO_CLOUD or set it to 0, make sure ~/.ollama/server.json does not contain "disable_ollama_cloud": true, restart Ollama, then sign in with ollama signin so it can fetch remote model details.

Why was cloud disabled in the first place?

OLLAMA_NO_CLOUD=1 or the app's Cloud switch is commonly turned off for offline or privacy-sensitive setups. If that's intentional, keep it and use local models rather than referencing remote ones.

I set OLLAMA_NO_CLOUD=0 and it still fails. Why?

Ollama treats cloud as disabled if either source says so. The environment variable and disable_ollama_cloud in ~/.ollama/server.json are checked separately, so a true value in server.json keeps cloud off even when the variable is 0. Fix the file (or the app toggle) and restart.

Stop firefighting agent errors

Decoding errors one at a time is the manual version of what BetterClaw automates. Run your agents on a no-code AI agent platform with managed models, retries and config validation built in.

Deploy Your Agent — Start FreeBook a Demo

Free plan available · Pro $49/mo · BYOK · 7-day money-back guarantee