LLMstudio CLI
The CLI starts a language-model proxy and a request tracking service for debugging applications across model providers.
- 390 ★ on the repository
- Last commit
- MPL-2.0
Extended access
Why this level
- The proxy receives provider credentials and application prompts; review access to logs and network ports.
Install
In your terminal, with SkillFoxx CLI
npx skillfoxx add cli/llmstudioDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The proxy receives provider credentials and application prompts; review access to logs and network ports.
This is third-party code. Review the repository files before installing.
What it does
LLMstudio’s CLI starts local proxy and request tracking services. Application teams can route model calls through a shared layer and observe usage. The project also includes a Python SDK and a prompt playground. The proxy supports cloud providers and models served through Ollama. Routing and fallback connections provide another request path when the primary provider is unavailable. Developers can use the SDK to integrate those calls in application code.
Who it is for. LLM application engineers comparing providers and inspecting request history during debugging.
Good fit when
- Your application calls several models and needs one request entry point.
- During a prompt experiment, you inspect provider usage and behavior.
- The team is testing a cloud model alongside an Ollama model.
- For a resilience test, you check fallback to a secondary provider.
Not a fit when
- You need a ready MCP server with project tools.
- The user only needs a chat without setting up Python.
- Project policy forbids recording model request contents in a separate service.
Example request
Prepare a separate LLMstudio environment and show how to route application test calls through the proxy with request tracking.Limitations
The full install includes proxy and tracker dependencies; the core package alone does not provide the same setup. Cloud models require your own provider keys and requests remain subject to each provider’s billing terms. The launch example exposes services on 0.0.0.0, so restrict network access before running it on a shared machine. Type casting is listed as a future feature and should not be treated as an available workflow dependency.
How to disable. Stop the proxy and tracker processes, then restore the application’s previous provider endpoint. Remove unused credentials from the environment.
FAQ
How do I start both services?
Run `llmstudio server --proxy --tracker` after installing the full package and configuring keys.
Where are the API methods documented?
The proxy serves documentation on port 50001 and the tracker on 50002, both at `/docs`. Open the appropriate address on the service host.
Can I connect my own model?
Yes, the project supports models through Ollama. Configure that provider and verify it with a separate test call.
Which package should I use without the proxy?
Use `pip install llmstudio` for the lightweight core when the additional services are unnecessary.
Where should provider keys go?
Place them in `.env` at the service launch location. Add only the needed keys yourself and keep the file out of published sources.
Discussion
No questions or reviews yet.
Related
CLI toolsA local Rust proxy that compresses the model API requests of Claude Code, Codex and Cursor so you pay for fewer tokens
A self-hosted gateway for multiple LLM keys and providers: rotation, failover, request logs and usage accounting
A CLI runs a model and MCP-tool gateway with owned API keys, format translation, routing and request logs.
A local proxy between AI agents and model providers that prices every run as it happens and cuts off spend above your cap
A local bridge from an MCP client to IAM-protected MCP servers on AWS: it signs requests with SigV4 using your AWS credentials
A proxy between MCP transports: connects stdio clients to remote SSE and HTTP servers and exposes local servers over the network
README in short
The package includes LangChain integration and batch calls. Repository notebooks provide usage examples.