OpenLLM vs LiteLLM
LiteLLM is a large open-source Python proxy with virtual keys, budgets and 100+ providers behind one OpenAI-compatible endpoint. It also documents subscription access: ChatGPT through an OAuth device flow, and Claude Max by forwarding each user's OAuth token. LiteLLM for teams operating a gateway; OpenLLM for the developer who wants their plans behind one key with nothing to host.
curl -fsSL https://www.openllm.sh/install | bashopenllm claudeOpenLLM vs LiteLLM
| Feature | OpenLLM | LiteLLM |
|---|---|---|
| Shape | Hosted gateway plus a small local daemon; single-user today | Self-hosted Python server with team features (budgets, RBAC) |
| Subscriptions | Claude, ChatGPT, Kimi Code, Grok and Cursor plans, served by a daemon on your machine that runs the vendor's own login | ChatGPT via OAuth device flow on the server; Claude Max by forwarding the user's token |
| Fallback | Per-alias chains shared by every client; a rate-limited entry is cooled and the chain continues before any output | Router retries and fallbacks per deployment |
| Coding-agent setup | openllm claude, openllm codex, openllm opencode: a per-session overlay that leaves the client's config untouched | Point each client at your proxy's base URL |
| Code search & memory | Built-in MCP server: semantic code search, doc indexing, memory | Not included |
| Self-hosting, local models, Windows | No | Yes |
| Credentials | API keys encrypted at rest under your vault, decrypted in memory only while serving a request | Held by your proxy deployment |
| Price | Free plan with a monthly token allowance, no card; no markup on provider tokens | Open source; you host it |
Pick LiteLLM when
- You run a team and need per-user budgets, RBAC or your own infrastructure
- You need local models or Windows
- You already operate it and it works
Pick OpenLLM when
- You want one key for Claude Code, Codex and OpenCode without running a server
- You want subscription logins to stay on your own machine
- You are one developer, not a platform team
Your subscriptions stay on your machine
Subscription providers run through the local OpenLLM daemon, which delegates to the vendor's official CLI and its own login. The hosted gateway never holds a subscription credential.
Provider keys encrypted at rest
API keys are encrypted in your browser with AES-256-GCM under a key derived from your recovery phrase, and stored as ciphertext. They are decrypted in memory only while a request is being served, then discarded. Subscription credentials never reach the hosted service.
Fallback chains when a model is rate-limited
Define an ordered chain per model alias. When a hop returns a rate limit, an overload or a provider error, the request moves to the next hop instead of failing your session.
- Chains are configured once and shared by every client and device
- Each hop resolves its own model settings, so a fallback never inherits the wrong parameters
- Every hop that ran is visible in your request history
Code search and memory over MCP
The same binary ships one MCP server: semantic search over your indexed repositories and documentation, plus memory that persists across sessions and clients.
Questions
Something else? Contact us
Checked 2026-09-25 against LiteLLM ChatGPT subscription provider, LiteLLM Claude Code Max tutorial.