API guide
Every model is called the same way: submit an asynchronous task, poll it, download the result. Base URL: https://api.hermes-ai.net
How it works
Submit
POST /api/v1/jobsSend a model slug and its input. You get a task ID right away.
Poll
GET /api/v1/jobs/{id}Check every few seconds until the task succeeds or fails.
Download
GET /api/v1/results/{id}Fetch each result with the same credentials.
Get started
API reference
Authentication
Create API keys, send them with each request, and understand scopes and endpoints.
ReadChoose a model
Find a model, read its input schema and send only the fields it accepts.
ReadTasks and results
Submit asynchronous tasks safely with idempotency, upload media, poll for status and download results.
ReadChat completions (LLM)
Call LLMs through an OpenAI-compatible endpoint with per-token billing from prepaid credits.
ReadErrors
Every error code the API returns, what it means and what to do next.
ReadIntegrations
OpenAI SDK
Point any OpenAI SDK at the Hermes base URL and call LLMs with your Hermes API key.
ReadClaude Code
Add the Hermes AI API MCP server to Claude Code, sign in with OAuth and optionally install the Agent Skill.
ReadCodex
Add the Hermes AI API MCP server to the Codex CLI and sign in from the terminal.
ReadCursor
Add the Hermes AI API MCP server to Cursor's mcp.json and connect with OAuth.
ReadVS Code
Add the Hermes AI API MCP server to .vscode/mcp.json and sign in from VS Code with OAuth.
ReadClaude
Add Hermes AI API to Claude as a custom connector with the MCP server URL and sign in with OAuth.
Read