Back to blog

Tools & integrationsGuide

How to connect Hermes Agent to CheapTokens

Connect Hermes Agent by Nous Research to CheapTokens as a custom OpenAI-compatible endpoint — with hermes model or config.yaml — and check that it answers.

By CheapTokens TeamPublished 1 min read

Hermes Agent connected to the CheapTokens API
On this page
  1. Before you start
  2. Option 1. Interactive setup
  3. Option 2. Edit config.yaml
  4. Check that it works

Hermes Agent from Nous Research supports any OpenAI-compatible API as a custom endpoint. Give it the CheapTokens address, key and a model, and the agent runs on open-weight models at fixed per-token prices. The steps follow the Hermes Agent documentation (Quickstart, AI Providers, Configuration) as of October 15, 2026; they have not been run on a live installation yet.

Before you start#

  • Hermes Agent is installed.
  • You have a CheapTokens API key and a positive balance — see Your first API request.

Option 1. Interactive setup#

Run hermes model, choose Custom endpoint (self-hosted / VLLM / etc.) and enter:

  • API base URL: https://api.cheaptokens.dev/v1
  • API key: your CheapTokens key
  • model name: deepseek-ai/DeepSeek-V4-Flash-0731

Hermes saves the endpoint in ~/.hermes/config.yaml.

Option 2. Edit config.yaml#

yaml
model:
  default: deepseek-ai/DeepSeek-V4-Flash-0731
  provider: custom
  base_url: https://api.cheaptokens.dev/v1

The key can stay in ~/.hermes/.env rather than in the YAML; the interactive setup asks for it. Hermes needs a large context for agent work; our models have 200K–390K tokens. Other models: zai-org/GLM-5.3-Flash, MiniMaxAI/MiniMax-M2.7 — see GLM vs DeepSeek vs MiniMax.

Check that it works#

Start a chat and ask a short question. If Hermes opens but replies come back empty or broken, run hermes model again and check the address, the model name and the key. Status codes are explained in Connection errors and fixes.

By default Hermes also sends side tasks — context compression, titles, image analysis — to your main model, so they are billed too. How to estimate an agent's spend: What an AI agent costs.

Tags: