Reduces token usage from paid providers by offloading work to local LM Studio models. Use when: (1) Cutting costs—use local models for summarization, extraction, classification, rewriting, first-pass review, brainstorming when quality suffices, (2) Avoiding paid API calls for high-volume or repetitive tasks, (3) No extra model configuration—JIT loading and REST API work with existing LM Studio setup, (4) Local-only or privacy-sensitive work. Requires LM Studio 0.4+ with server (default :1234). No CLI required.
Changelog **New helper scripts for load/unload/test added; unload guidance clarified** - Added scripts: load.mjs (model loader), unload.mjs (unloads all instances for model key), test.mjs (basic test suite). - Significantly expanded and clarified unload instructions: always use instance_id from GET /api/v1/models (never assume it matches model key; keys can have multiple instances with ids like key:2). - Improved example workflows, especially for multi-turn and unload/verify unload steps. - Documented new script features (like --unload-after, --keep, --stateful, --log). - Added troubleshooting and error handling procedures for instance management. - Updated copy-paste and example sections for practical ease-of-use.