Give LM Studio a few new superpowers. Discover plugins and MCP servers for the things you want to make, find, and do.
Native plugins can be added directly to LM Studio. For MCP servers, “Ready” means we can generate a configuration from the published metadata. It is not a security review. Check the tool and its permissions before adding it.
Never run out of context in LM Studio: threshold-driven compaction, mid-task compression for agentic loops, tool-result capping, and Agentic Workspace state preservation.
Keeps a long chat going past the context limit, in the same window. Summarizes older messages into a consolidated state and sends the model that instead of the full history; your chat is untouched and every message stays on screen. Tool results are capped so one large call cannot kill a reply. Optionally archives transcripts and state notes to an Obsidian vault. Read 'When not to use it' first: it suits long working conversations, not bulk data analysis.
Inspect, trim, convert, and manage FFmpeg video jobs locally from LM Studio.
Converts attached videos into image frames with ffmpeg before they reach the model, so any vision-capable LLM can 'watch' them. Requires ffmpeg/ffprobe and a vision model. Version 1.1 Changelog: v1.1 - Persistent frame storage (default): frames are copied to LM Studio's user-files cache so follow-up questions about the same video keep working across turns/restarts; per-chat registry tracks files. New /video commands: list, on <name>, off [name], purge <name|all>. Max 'Frame count' and 'Max frames per video' raised from 100/400 to 1000. v1.0 - Initial release: ffmpeg frame extraction (fixed-FPS or fixed-count sampling, resolution cap) for vision models.
Local toolbox for files, live web, research, screenshots, audio, and video.
A web tools plugin for LM Studio built on the impit client for improved bot detection evasion. Includes a Visit Website tool and Web Search backed by DuckDuckGo. An LM Studio plugin that gives local LLMs four web-oriented capabilities: searching the web, searching for images, visiting a page and reading its content, and downloading images into the chat's working directory so the assistant can embed them inline.
Version 5.10.0
LM Studio plugin and local stdio MCP entrypoint made for Bionic, living side-by-side in the `generate-image` LM Studio plugin repo. Provides agentic image and video generation via a local Draw Things instance.
No description provided on LM Studio Hub.
complete Web Search
Fork of Altra web-search - Fact-grounded web search with source verification — searches the web, reads full pages, fact-checks claims, and flags single-source info as unverified. No API key required. Fixes oriented on stability and reliability - especially around default DDG/Bing scrapers.
An LM Studio plugin that turns LM Studio into a client for any OpenAI-compatible API. It is aimed chiefly at the Splash local model runner — a high-throughput, Apple-Silicon, OpenAI-compatible inference server.
All-in-one LM Studio plugin: 50+ local tools, persistent memory, sub-agent delegation, design systems, and media analysis. Modular architecture with smart output caps on all tools, near-duplicate memory detection (TF-IDF + Jaccard), HTML asset auditing, tiered memory injection (L0 identity → L1 essential → L2 contextual), 58 pre-cached design system references, pattern-matched delegation hints (EN/PT), and auto hints that guide local models to use efficient editing patterns. Optimized for models with limited context (4K-20K tokens).
This plugin enables large language models in LM Studio to perform file and directory operations through a set of safe, controlled tools. It bridges the gap between AI assistants and the file system while maintaining security through path traversal protection.
No description provided on LM Studio Hub.
Give your local model web search. DuckDuckGo backend works out of the box (no API key); SearXNG backend lets you point at your own privacy-respecting metasearch instance (Google Bing Brave DuckDuckGo combined).
No description provided on LM Studio Hub.
No description provided on LM Studio Hub.
Next-level image search and information retrieval based on multimodal embeddings. find-image supports text-based, image-based, and metadata-based queries using Qwen3-VL-Embedding.
Autonomous coding agent for LM Studio: plans, edits transactionally with rollback, runs commands, uses Git/GitHub, researches the web, runs sub-agents. Manual/Plan/Auto approval.
Version 1.0.2
No description provided on LM Studio Hub.
An LM Studio plugin for web search with multiple providers, webpage reading, image search, access current date & time and more.
Version 1.0.4
