GitHunt
GD

gdmka/openai-compat-endpoint

Chat with cloud AI models directly from LM Studio!

Table of Contents

OpenAI-Compatible Client for LM Studio

Turn LM Studio into a cloud-ready powerhouse. Access local and cloud inference simultaneously in the same app!
This fork lets you chat with Cerebras, Groq, OpenRouter, Claude, GPT-4o, Gemini-2.5, DeepSeek, Kimi-K2, GLM V4.5—or any OpenAI-shaped endpoint—without ever leaving the comfy LM Studio UI.

What’s new

  • Added support for reasoning effort and thinking for OpenAI and Anthropic compatible models
  • Native OpenAI sampling knobs (temperature, top-p, frequency penalty, etc.) exposed in the GUI
  • Zero config: paste your API key, pick the model, start vibing
  • Keeps all local-model superpowers intact—switch between cloud and local on the fly
  • Added custom system prompt support

Install in 5 seconds

  1. Navigate to LM Studio Hub
  2. Hit “Run in LM Studio” on the plugin page
image
  1. Done—plugin will be available in Chat view under Your Generators section
image

Unlock the full power of productivity by instantly navigating between search terms with built-in full-text search across your chat history.

image

Quick start

  1. Grab an API key from your favorite provider (Cerebras, Groq, OpenRouter, Anthropic, OpenAI, etc.)
  2. In LM Studio → Chat → Your Generators load the plugin → hit Show Settings shortcut → pick Generators tab → paste key
  3. Type model name, select AI provider, tweak sampling, chat away!
image

Roadmap & bugs

Got feature requests or bugs?
Drop them in the Issues tab—every ticket gets love.

Support

If you find this project helpful, consider buying me a coffee!

Buy Me A Coffee

Project Supporters

@jrdubbleu