# LM Studio

**Canonical:** https://apis.io/providers/lm-studio/  
**Website:** https://lmstudio.ai  
**APIs profiled:** 4

LM Studio is a desktop application for discovering, downloading, and running large language models (Llama, DeepSeek, Phi, Qwen, gpt-oss, and others) locally on macOS, Windows, and Linux. It runs GGUF and MLX models via llama.cpp and Apple's MLX engine, and ships a local HTTP server that developers can drive from their own apps: a native LM Studio REST API (beta) under /api/v1 with generation stats, model load/unload/download and MCP-via-API, an OpenAI-compatible surface under /v1 (chat/completions, completions, embeddings, models, responses), and an Anthropic-compatible /v1/messages endpoint. It is complemented by Python (lmstudio) and TypeScript (@lmstudio/sdk) SDKs, the MIT-licensed lms command-line tool, headless/daemon operation, LM Link cross-device routing, and MCP host support. Backed by Matrix Partners.

## Kin Score — 29.6 / 100 (thin)

Scored 2026-08-20 under rubric 0.12.0. Trend: flat (+0.0 from 29.6).

| Facet | Score |
|---|---|
| Discoverability | 72.2 |
| Contract Quality | 12.2 |
| Governance | 16.7 |
| Contract Governance | 16.7 |
| Operational Transparency | 2.6 |
| Developer Ergonomics | 63.7 |
| Commercial Clarity | 21.1 |
| Access Clarity | 21.1 |

## Agent readiness — 28.4 (agent-aware)

| Dimension | Value |
|---|---|
| Spec Presence | yes |
| Agentic Access | no |
| Reversibility Documented | no |
| MCP Server | no |
| Auth Clarity | yes |
| Idempotency | no |
| Error Semantics | documented |
| OpenAPI Examples | no |
| Rate Limit Signal | no |
| Event Surface Described | no |
| Agent Skills | derived |
| Well Known Catalog | no |
| Consent Identity | no |
| Agent Card | no |
| Dry Run Mode | no |

## Access

Self-serve signup — onboarding: self-serve, pricing: unknown, trial: no (confidence: medium).

## APIs (4)

- **LM Studio Anthropic Compatibility API** — Anthropic-compatible messages endpoint.
- **LM Studio Chat API** — Native LM Studio chat inference with generation stats.
- **LM Studio Models API** — List, load, unload, and download local models.
- **LM Studio OpenAI Compatibility API** — OpenAI-compatible endpoints for reuse of existing OpenAI clients.

## MCP servers (1)

- **LM Studio MCP Server**

## Security (3)

- **Lm Studio Authentication** — http · 1 scheme
- **Lm Studio Domain Security** — TLSv1.3 · DNSSEC · DMARC
- **Lm Studio Vulnerability Disclosure** — disclosure policy published

## Tags

Company, Artificial Intelligence, Local LLM, Machine-Learning, Inference, Developer Tools, LLM, MCP, SDK, Desktop Application

---

Profiled by [API Evangelist](https://apievangelist.com) and published on [APIs.io](https://apis.io/providers/lm-studio/). Scores are computed from the provider's own public artifacts under a published rubric.