A local routing layer that lets coding agents use subscriptions, API keys and local models through one key, with fallback and shared memory.
OpenLLM homepage
1 / 3Overview
What is OpenLLM?
Take your AI tokens anywhere
OpenLLM is a local routing layer for developers who pay for several AI subscriptions and API keys and run local models, but find each one stuck in its own app. Connect them once and get one key that works in Claude Code, Codex, Kimi Code, Grok, Cursor and Muse Code, plus any app that accepts an OpenAI or Anthropic key.
The layer adds automatic fallback across providers, shared memory across agents, web data through Context.dev, and usage and cost analytics. Provider keys are encrypted in the browser under a recovery phrase only the user holds. The Mac app and a terminal setup are available, and OpenLLM works on macOS and Linux, with a web app and remote sessions.
The free plan costs $0, includes inference and 50M tokens a month across 2 devices, with no card required. Pro costs $30 per month for up to 5 devices, with unlimited tokens and a 3-day free trial.
Platforms and languages
OpenLLM Availability
Platforms
Languages
Capabilities
OpenLLM Key Features
One key for every model
Connect subscriptions, API keys and local models once, then use them in Claude Code, Codex, OpenCode and other tools through one key.
Automatic fallback
Put models in order and the next one picks up when a limit is hit, so work keeps going.
Shared memory
Save a decision or preference once and every connected app can recall it, on every plan.
Included inference
Every account comes with inference included, so the first request works before anything is connected.
Usage and cost analytics
See which models answered and how many tokens were used, across providers and tools.
Best for
Who uses OpenLLM?
Developers using coding agents such as Claude Code and Codex
Use all subscriptions, API keys and local models through one key
Plans and access
OpenLLM Pricing
Freemium
Free $0 with 50M tokens/month; Pro $30/month with a 3-day free trial
Common questions
OpenLLM FAQs
Is there a free plan?
Yes. The free plan costs $0, includes inference, and covers 50M tokens a month across 2 devices, with no card required.
What happens when I hit a model limit?
OpenLLM falls back to the next model in your order, so your coding agent keeps working.
Which tools does OpenLLM work with?
OpenLLM works with Claude Code, Codex, OpenCode, Kimi Code, Grok, Cursor and Muse Code, plus API keys, custom keys and local models.
Is there a Pro trial?
Yes. Pro costs $30 per month for up to 5 devices and includes a 3-day free trial.
Reviews
See what the community thinks and share your experience.
—
Based on 0 ratings
Rating distribution
Community reviews
No reviews
No written reviews yet.
