---
title: "Use RunAPI in Hermes Agent — Custom Endpoint Setup Guide"
url: "https://runapi.ai/hermes-agent.md"
canonical: "https://runapi.ai/hermes-agent"
locale: "en"
---

# RunAPI in Hermes Agent

Hermes Agent can use a custom OpenAI-compatible endpoint for chat models. Paste the setup prompt and Hermes creates the provider — key goes in RUNAPI_API_KEY, base URL points to /v1, and run hermes doctor before a long session.

## Install prompt

Copy this prompt into Hermes Agent:

```text
Configure this Hermes Agent install to use RunAPI.

Requirements:
- Read the API key from RUNAPI_API_KEY. Do not hardcode the key.
- Add a custom provider named runapi.
- Use base_url "https://runapi.ai/v1".
- Use key_env RUNAPI_API_KEY.
- Use api_mode chat_completions.
- Set model.provider to custom:runapi.
- Set the default model to gpt-5.5.
- Run `hermes doctor` and report any config problems before starting a session.
- If RUNAPI_API_KEY is missing, ask me to set it before starting Hermes Agent.
```

## Configuration

```yaml
custom_providers:
  - name: runapi
    base_url: "https://runapi.ai/v1"
    key_env: RUNAPI_API_KEY
    api_mode: chat_completions

model:
  provider: custom:runapi
  default: gpt-5.5
```

## Steps

1. **Paste the setup prompt** — Copy the prompt from the first tab into Hermes Agent. It tells Hermes Agent to read RUNAPI_API_KEY and add RunAPI as a named custom provider.
2. **Review the generated config** — Hermes Agent should set provider to custom:runapi and default to gpt-5.5, with key_env pointing at RUNAPI_API_KEY.
3. **Run diagnostics** — Run hermes doctor after editing config, then open Hermes Agent and send a short test prompt before using it for a larger task.

## FAQ

### Can Hermes Agent use RunAPI as a custom endpoint?

Yes. Hermes Agent supports model selection and custom endpoints. Configure the named custom provider with the RunAPI /v1 base URL, keep the API key in RUNAPI_API_KEY, and select a RunAPI model such as gpt-5.5.

### Should I use hermes model or edit the config file?

Use hermes model when you want the guided picker. Edit the config file when you need a repeatable setup for a server or shared environment. In both cases, verify with hermes doctor afterward.

### Where should RUNAPI_API_KEY be set for Hermes Agent?

Set RUNAPI_API_KEY in the same shell, service unit, or gateway environment that launches Hermes Agent. GUI or daemon launches do not always inherit your interactive shell variables.

### Can Hermes Agent still use other providers after adding RunAPI?

Yes. Adding RunAPI as custom:runapi does not remove other Hermes Agent model options. Use hermes model or the /model command when you want to switch between configured models.

### Can Hermes Agent generate images through RunAPI instead of ComfyUI?

Yes. RunAPI exposes image models like Flux Kontext, Imagen 4, and GPT Image through the same /v1 endpoint. You do not need a ComfyUI cloud instance or GPU rental — Hermes Agent calls RunAPI directly and receives image URLs in the response.

### Does RunAPI support music generation from Hermes Agent?

Yes. RunAPI provides direct access to Suno v4 through v5.5 for music generation. The same RUNAPI_API_KEY and custom:runapi provider handle music requests alongside chat, image, and video — no separate bridge service or API key required.


## Links

- [Model catalog](https://runapi.ai/models)
- [CLI](https://runapi.ai/cli)
- [MCP Server](https://runapi.ai/mcp)
- [Pricing](https://runapi.ai/pricing)
- [API docs](https://runapi.ai/docs)
- [OpenClaw setup](https://runapi.ai/openclaw)