Integrations
UberLLM works in any app that lets you set an OpenAI-compatible base URL, and in apps built for the Anthropic API. Every guide below was checked against the app's current settings screens and tested through the live API on 2026-10-30.
- OpenAI-compatible base URL:
https://api.uberllm.dev/v1 - Anthropic-compatible base URL:
https://api.uberllm.dev(/v1/messages) - API key: create one in Dashboard → Keys. Keys
start with
ull_. - Model ids: the catalog, or
GET /v1/modelswith your key. - Want today's best models for your app, with prices, and a config that names them? See Set up your tool (also covers the OpenAI Python SDK).
No UberLLM fee on requests. Adding credit by card costs 5.5% + $0.50.
Coding agents and editors
| App | How it connects |
|---|---|
| Claude Code | Anthropic base URL; one install command |
| Codex CLI | Custom model provider, wire_api = "responses" |
| Cline | OpenAI Compatible provider |
| Kilo Code | Custom provider, or kilo.jsonc |
| OpenCode | /connect + opencode.json |
| Cursor | OpenAI key + base URL override (chat only) |
| Goose | Custom provider JSON file |
| Roo Code · Aider · Continue | OpenAI-compatible settings |
Chat apps
| App | How it connects |
|---|---|
| Open WebUI | OpenAI API connection, or a direct connection per user |
| LibreChat | Custom endpoint in librechat.yaml |
| Raycast | Custom provider in providers.yaml (Raycast Pro) |
| Chatbox | OpenAI API Compatible provider, or a one-click import link |
Roleplay apps
| App | How it connects |
|---|---|
| SillyTavern | Custom (OpenAI-compatible) chat completion source |
| JanitorAI | Proxy configuration with the full chat URL |
| RisuAI | Custom API model |
Automation
| App | How it connects |
|---|---|
| n8n | OpenAI credential with a Base URL, OpenAI Chat Model node |
| n8n video templates | Five workflows to import by URL: Veo 3.1, Kling, faceless Shorts, UGC ads, Gemini Veo drop-in |
| ComfyUI | Video and image nodes; key from UBERLLM_API_KEY |
SDKs and agents
- OpenAI SDK · Anthropic SDK
- Vercel AI SDK · LangChain
- MCP server: UberLLM's models as tools in Claude Desktop, Cursor and other MCP clients.
- Coding agent setup in one line: tell your agent "Read https://uberllm.dev/SKILL.md and set me up with UberLLM."
Tell UberLLM which app sent a request
Where an app lets you add headers, add X-Title: <app name> (and, if you like,
HTTP-Referer: <app URL>). Each guide shows where.
Your app is missing?
If it has a field for an OpenAI-compatible base URL, it works: use the values at the top of this page. If something fails, write to [email protected] with the app name and the error.