Integrations

Most AI clients speak OpenAI or Anthropic. Melious implements both, so nearly every tool in the ecosystem works against us by changing a base URL. For a handful of coding agents we go further — the Melious CLI installs them, writes the config, and launches them with the environment already set.

If you came looking for the OpenAI or Anthropic SDK itself, the migration quickstarts live in get started — "switching SDK" is a different task than "integrating a tool", and the integrations section is for clients built on top of those SDKs.

Coding agents

Agents that run in your terminal and edit code. Claude Code, OpenCode, Mistral Vibe, Pi, and Codex are installable through the Melious CLI — melious tools install <name> installs the tool, writes the config, and launches it with the environment already set. Aider and Hermes Agent you wire by hand.

• Claude Code — Anthropic's coding agent, routed through the Anthropic-shape endpoint
• OpenCode — open source, token-efficient, pre-configured by the CLI
• Mistral Vibe — Mistral's terminal agent, pointed at Melious via its OpenAI provider
• Pi — minimal terminal harness; wire Melious in through a TypeScript extension
• Codex CLI — OpenAI's terminal agent; a TOML provider block against our Responses endpoint
• Hermes Agent — Nous Research's self-improving agent; a drop-in provider profile, and it runs their Hermes 4 models
• Aider — AI pair programming in your terminal; two env vars and an openai/ model prefix

IDEs

• Cursor — paste the Melious base URL into Cursor's OpenAI-compatible settings
• Continue.dev — VS Code and JetBrains assistant with a Melious provider config
• Cline — autonomous agent in the VS Code sidebar, wired in through its OpenAI-compatible provider

SDKs and frameworks

The OpenAI and Anthropic SDKs themselves are covered in get started. These integrations sit one layer up.

• Vercel AI SDK — createOpenAI with our base URL, and every framework primitive works
• LangChain — ChatOpenAI from langchain-openai inherits every upstream update
• LlamaIndex — RAG pipelines against Melious models and (optionally) embeddings
• LiteLLM — the AI gateway a lot of the ecosystem already runs on; one config block covers both our API shapes

Chat interfaces

• LibreChat — self-hosted chat UI; one YAML block to add Melious as an endpoint
• Claude Desktop — Anthropic's desktop app, pointed at Melious through its third-party gateway mode
• LobeChat — self-hosted chat UI; three env vars on the container point its OpenAI provider at us
• Open WebUI — self-hosted chat UI for teams; two Docker env vars and the model picker fills itself

Automation

• n8n — workflow automation; use the OpenAI Chat Model node with Melious' base URL
• OpenClaw — self-hosted gateway bridging chat apps to coding agents; add Melious as a custom provider

What you won't find here

A tool isn't listed because (a) we haven't written a guide for it yet, or (b) it doesn't speak OpenAI or Anthropic shapes yet. If your tool does speak one of those shapes, it'll likely work anyway — drop us a line at support@melious.ai and we'll add it.