Using TOKENOPEN API with Hermes Agent

Use Nous Research's Hermes Agent with TOKENOPEN API to build an autonomous AI assistant with persistent memory

Hermes Agent is an open-source autonomous AI Agent from Nous Research. It features persistent memory, multi-platform integration (20+ messaging platforms), and long-context processing — ideal for deploying on servers for continuous tasks. By configuring TOKENOPEN API, you can use all models available on the platform.

Hermes Agent requires a model with a 64K+ token context window. Choose a long-context model such as gpt-5.5 or claude-sonnet-4-6.

Step 1: Install Hermes Agent

Use the official install script:

curl -fsSL https://raw.githubusercontent.com/NousResearch/hermes-agent/main/scripts/install.sh | bash

After installation, run the post-install step:

hermes postinstall

Run in PowerShell:

iex (irm https://raw.githubusercontent.com/NousResearch/hermes-agent/main/scripts/install.ps1)

After installation, run the post-install step:

hermes postinstall

Install via pip:

pip install hermes-agent

After installation, run the post-install step:

hermes postinstall

Verify the installation:

hermes --version

Step 2: Configure TOKENOPEN API

Get an API Key

Configure the API Endpoint

Write the TOKENOPEN API configuration using hermes config commands:

hermes config set model.provider custom

hermes config set model.base_url "https://www.tokenopen.ai/v1"

hermes config set model.api_key "replace-with-your-api-key"

hermes config set model.default "gpt-5.5"

To manually edit the config file at ~/.hermes/config:

vim ~/.hermes/config

Write the TOKENOPEN API configuration using hermes config commands:

hermes config set model.provider custom

hermes config set model.base_url "https://www.tokenopen.ai/v1"

hermes config set model.api_key "replace-with-your-api-key"

hermes config set model.default "gpt-5.5"

To manually edit the config file at ~/.hermes/config, open it in TextEdit:

open -e ~/.hermes/config

Run the following commands in PowerShell:

hermes config set model.provider custom

hermes config set model.base_url "https://www.tokenopen.ai/v1"

hermes config set model.api_key "replace-with-your-api-key"

hermes config set model.default "gpt-5.5"

To manually edit the config file at %USERPROFILE%\.hermes\config:

notepad "$env:USERPROFILE\.hermes\config"

model.base_url must end with /v1. Set model.api_key to the key obtained from the TOKENOPEN Console.

Or Configure via Interactive Wizard

Run hermes model to enter the model selection wizard and choose Custom Endpoint:

hermes model

In the wizard:

Provider: select Custom Endpoint
Base URL: enter https://www.tokenopen.ai/v1
API Key: enter your TOKENOPEN API Key
Model Name: enter gpt-5.5 or another desired model ID

Step 3: Use Hermes Agent

Launch Hermes

Classic terminal mode:

hermes

Launch Hermes

Modern TUI interface (recommended):

hermes --tui

The TUI mode provides a friendlier interface with history, multi-session switching, and more.

Run Initial Setup

On first use, run the setup wizard to configure basic preferences:

hermes setup

Connect to Messaging Platforms (optional)

Hermes Agent supports integration with many messaging platforms (Telegram, Discord, Slack, etc.). Refer to the official documentation for setup instructions.

Long-context models supported on TOKENOPEN (examples):

Model ID	Context	Description
`gpt-5.5`	1M	GPT-5.5, latest flagship model
`gpt-5.4`	1M	GPT-5.4, balanced performance & cost
`gpt-5.4-mini`	400K	GPT-5.4 Mini, low-latency lightweight model
`claude-sonnet-4-6`	200K	Claude Sonnet 4.6, ultra-long context
`deepseek-r1`	128K	DeepSeek R1, strong reasoning

For the full model list and pricing, visit the Model Marketplace.

Common Commands

Command	Description
`hermes`	Launch in classic terminal mode
`hermes --tui`	Launch with TUI interface
`hermes setup`	Run the initial setup wizard
`hermes model`	Switch model and provider
`hermes config set [key] [value]`	Set a configuration value
`hermes postinstall`	Run post-installation setup

Using TOKENOPEN API with Hermes Agent

On this page