Continue extension not showing local Ollama models — config looks correct?

On April 1st, a developer reported that the Continue extension — the popular open-source AI coding assistant that plugs into VS Code and JetBrains — was failing to surface locally running Ollama models in its model picker, despite the configuration appearing correct. The frustration is a familiar one in the local AI tooling space: everything looks right on paper, the config file points where it should, Ollama is running, the models are pulled and available, and yet the integration stays silent. No error. No models. Just a dropdown that should be populated and isn't. The post attracted enough commiseration to suggest this isn't an isolated environment quirk, but not enough verified reproduction to call it a genuine regression.

If confirmed, here is what this means. Continue has quietly become one of the most important bridges between local LLM infrastructure and everyday developer workflows — it's how a lot of privacy-conscious engineers run Ollama without their code leaving their machine. A silent failure in model discovery doesn't just inconvenience; it erodes trust in the entire local-first stack at the exact moment that stack is trying to prove itself as a credible alternative to cloud-dependent tools. Developers who hit this and don't know to dig into logs or community threads will simply assume local models are too brittle and reach for a hosted API instead. The second-order effect is churn away from self-hosted setups, and that matters for the broader ecosystem.

Watch for a response from the Continue maintainers on GitHub confirming whether a recent update changed how the extension discovers Ollama endpoints, and whether other users can reproduce the issue across different operating systems and Ollama versions.

NewsHive monitors these sources continuously. All signal titles above link to the original reporting.

Intelligence by NewsHive. Need help navigating what this means for your business? Contact GeekyBee →