Choose a local summary model
Outcome
Section titled “Outcome”Local Mode has an installed Ollama model selected for summaries and chat.
- In Settings, choose Local (Ollama) as the AI provider.
- Open Models.
- Under Summarization & Chat, select an installed model or expand Download more models.
- Download a catalog model when needed, wait for the progress to finish, then select it as the local model.
- Regenerate a reviewed test summary to evaluate whether it meets your needs.
Current local-model examples
Section titled “Current local-model examples”| Capability | Status | Current behavior |
|---|---|---|
| Llama 3.2 3B | Available | Lightweight local model exposed by the curated Ollama catalog and recommendation heuristic. |
| Llama 3.1 8B | Available | Default local summary and chat model and the balanced recommendation for higher-memory machines. |
| Phi 3.5 3.8B | Available | Lightweight local model exposed by the curated catalog and lower-memory recommendation path. |
| Mistral 7B | Available | Optional local model exposed by the curated Ollama catalog. |
| Qwen 2.5 7B | Available | Optional local model exposed by the curated Ollama catalog. |
The full set of installed models is shown by the application. The curated catalog and estimated download sizes can change with the registry; neither is a recommendation of a particular model for every Mac or content type.
If Local Mode is unavailable
Section titled “If Local Mode is unavailable”When External API is selected, Model Manager does not use local Ollama models for summary and chat. Switch back to Local (Ollama) to manage them.