About this integration
Connect Claude to any OpenAI-compatible LLM endpoint and offload routine work to a local model.
- Transport
- stdio
- Authentication
- none
- Initial setup
- none
- Runtime
- unattended
- Evidence
- documented
- Version
- 3.2.3
- Package
- @houtini/lm
- Last compatibility test
- Not independently tested
Connect your agent
@houtini/lmThe reviewed default is a local LM Studio endpoint with no API key. Remote endpoints can optionally require a bearer token. Documentation was reviewed from pinned captures; the package and model endpoint were not executed.
Capabilities: Delegate bounded text and code tasks to OpenAI-compatible model endpoints, Discover available local models, Create embeddings and report inference statistics
Connected profiles
Additional details
io.github.houtini-ai/lm
Source ↗ · Checked 2026-09-173.2.3
Source ↗ · Checked 2026-09-17CC0-1.0; package licenses are separate
Source ↗ · Checked 2026-09-17npm
Source ↗ · Checked 2026-09-17@houtini/lm
Source ↗ · Checked 2026-09-193.2.4
Source ↗ · Checked 2026-09-19Publisher documents an MCP client launching the package and serializing background requests to local model providers.
Source ↗ · Checked 2026-09-19The default localhost endpoint has no API key; HOUTINI_LM_API_KEY is optional for authenticated endpoints.
Source ↗ · Checked 2026-09-19Claude Desktop launches npx -y @houtini/lm and points it to the default localhost model endpoint.
Source ↗ · Checked 2026-09-19Publisher documentation reviewed; package and integration endpoint not executed or independently security-audited.
Source ↗ · Checked 2026-09-19