# Image generation: create images from a text prompt

> Pick an image backend such as FAL.ai, OpenAI, OpenRouter, xAI or Krea so the agent can generate images on request.

Canonical: https://hermeswatcher.com/tools/image-generation  
Kind: Provider setup · Skills, plugins & MCP  
Last reviewed: Oct 20, 2026  
Basis: Explained from the official Hermes Agent docs and repository (Nous Research). Steps follow the linked official sources.

## What it is for

The image generation tool lets Hermes turn a prompt into a picture during a conversation. FAL.ai is the main backend, with a default fast FLUX model and a menu of others that trade speed against cost. You can also route through OpenAI, OpenRouter, xAI, Krea, Meta's model API, or the Nous Tool Gateway if you have a paid Nous Portal subscription. You choose the backend and model interactively in `hermes tools`, and the choice is saved to config.yaml. It suits thumbnails, social graphics and quick concept images without leaving the chat.

## What you need

- Credentials for the active backend, for example FAL_KEY, OPENAI_API_KEY or KREA_API_KEY
- Or a paid Nous Portal subscription to use the Tool Gateway instead of your own FAL key

## Setting it up

1. Create an API key in your fal.ai dashboard and add FAL_KEY to your environment
2. Run `hermes tools`, select Image Generation, choose a backend, then pick a model
3. Or set it directly, for example `hermes config set image_gen.provider fal` and `hermes config set image_gen.model <model-id>`

## How to check it works

- `hermes tools` marks the active model as currently in use
- Confirm image_gen.provider and image_gen.model in config.yaml
- Set IMAGE_TOOLS_DEBUG=true and inspect the per-session debug log under ./logs/

## Limits

- Image editing only works on edit-capable models; text-to-image-only models reject image inputs
- Backend image URLs expire after hours or days, though Hermes caches files locally
- Unsupported parameters are silently dropped, and per-image backends do not record token usage

## Related workflows

- [Collect content ideas without losing the evidence](https://hermeswatcher.com/workflows/content-research-queue.md)

## Sources

- [Hermes Agent docs: Image Generation](https://hermes-agent.nousresearch.com/docs/user-guide/features/image-generation)
- [Hermes Agent docs: Features overview](https://hermes-agent.nousresearch.com/docs/user-guide/features/overview)

---

HermesWatcher is an independent guide. It is not affiliated with or endorsed by Nous Research. The [official Hermes Agent documentation](https://hermes-agent.nousresearch.com/docs) and [repository](https://github.com/NousResearch/hermes-agent) from Nous Research are the authority; each page here lists the official sources it was written from. If this site and the official docs disagree, trust the official docs. Index for agents: https://hermeswatcher.com/llms.txt