Right-Size the Local Model

Recover a local Ollama service configured for a model the host cannot safely serve.

A beginner AI Engineering challenge worth 3 points. Solve it hands-on in a real AI Engineering environment in your browser - no local setup, no fake shells.

The Challenge

Scenario: A developer changed the assistant to an unavailable 7B model, a 32K context, and four parallel requests. The shared CPU server now fails its smoke test.

Repair service.env to use the preloaded qwen2.5:0.5b, cap context at 4096 or less, and allow one request at a time. Run ./smoke-test.sh, then check.