Ollama: model not found, try pulling it first

The message

model "<name>" not found, try pulling it first

Parts in <angle brackets> change from run to run.

What it means

The API was asked for a model this machine does not have. Ollama's API never downloads on its own; the name must match a pulled model, tag included.

Why it happens

• The model was never pulled on this machine (or in this container).
• The tag differs: qwen3:8b and qwen3 are different names.
• The client talks to a different Ollama server than the one you pulled with (another port, host or container).

How to fix it

1. ollama list shows exactly what is installed; use that name in the request.
2. Pull it on the server the client uses: ollama pull <name>.

Related errors

• Ollama: pull model manifest: file does not exist