integrations/coding agent/Hermes

Hermes + AIgateway

Nous Research's self-hosted agent — one config block points it at any model.

Hermes is Nous Research's open-source autonomous agent: it lives on your server, keeps persistent memory, runs scheduled work, and reaches you over Telegram, Discord, or Slack. Its model section takes any OpenAI-compatible endpoint, so AIgateway drops in as the brain — and the agent can run on any of our 1000+ models without touching the agent itself.

Hermes homepage →

Setup

Three steps or fewer.

STEP 01

Install Hermes

One line on Linux / macOS / WSL2. Then run the setup wizard.

curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup

STEP 02

Point its model at AIgateway

Hermes reads ~/.hermes/config.yaml. Set provider: custom and Hermes calls the base_url directly with your key — every model in our catalog becomes a drop-in brain.

# ~/.hermes/config.yaml
model:
  default: moonshot/kimi-k2.6
  provider: custom
  base_url: https://api.aigateway.sh/v1
  api_key: sk-aig-...

STEP 03

Swap models without touching the agent

Change default to any provider/slug from the AIgateway catalog — anthropic/claude-opus-4.7, openai/gpt-5.4, moonshot/kimi-k2.6. Same agent, same memory, different model.

Notes

provider: custom is the key — it tells Hermes to call any OpenAI-compatible base_url directly with api_key (or the OPENAI_API_KEY env var).
For an always-on autonomous agent, mint a sub-account key from AIgateway so its long-running spend lands under its own cap and shows up tagged in your dashboard.
Tool calling, streaming, and structured output pass through unchanged — AIgateway forwards the full OpenAI schema.

Hermes + AIgateway

Three steps or fewer.

Install Hermes

Point its model at AIgateway

Swap models without touching the agent

Same key. Every other tool.