Agents
Setup guide

Hermes Agent with an uncensored model

To run Nous Research's Hermes Agent on an uncensored model, add a custom provider to ~/.hermes/config.yaml with the API's base URL and the abliterated model ids. The config below lets Hermes run up to seven subagents at once.

Updated
FormatOpenAI Chat Completions
Endpoint/chat/completions
Configured in~/.hermes/config.yaml
Switch with/model and /reasoning
Context1,048,576

Which model to use in Hermes Agent

Change default and default_model to start on a model other than DeepSeek V4.1 Flash 1017.

ModelUse it forModel id
Kimi K3.1 ProHard multi-file and security workkimi-k3.1-pro-uncensored
DeepSeek V4.1 FlashEveryday work, and the default heredeepseek-v4.1-flash-1017-uncensored
GLM-5.3A second opinion, text onlyglm-5.3-uncensored
GLM-5.3 FlashBulk edits and many subagentsglm-5.3-flash-uncensored

Set it up

  1. Install Hermes Agent
    Shell
    curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
    PowerShell
    iex (irm https://hermes-agent.nousresearch.com/install.ps1)
  2. Set your API key

    Create a key at Silk Compute, export it, and add the line to ~/.zshrc or ~/.bashrc.

    Shell
    export SILK_API_KEY="sk-..."

    Hermes also reads SILK_API_KEY from ~/.hermes/.env.

  3. Add the provider

    Add this to ~/.hermes/config.yaml, replacing its model section. If it already has providers or delegation, merge these into them.

    ~/.hermes/config.yaml
    model:
      provider: custom:silkcompute
      default: deepseek-v4.1-flash-1017-uncensored
    
    providers:
      silkcompute:
        name: Silk Compute
        api: https://api.silkcompute.com
        key_env: SILK_API_KEY
        transport: chat_completions
        default_model: deepseek-v4.1-flash-1017-uncensored
        models:
          kimi-k3.1-pro-uncensored: {context_length: 1048576}
          deepseek-v4.1-flash-1017-uncensored: {context_length: 1048576}
          deepseek-v4.1-flash-1017-fast-uncensored: {context_length: 1048576}
          glm-5.3-uncensored: {context_length: 1048576}
          glm-5.3-flash-uncensored: {context_length: 1048576}
          hy4-uncensored: {context_length: 1048576}
    
    delegation:
      max_concurrent_children: 7
  4. Run it
    Shell
    hermes

    Switch models with /model custom:silkcompute:<model id> and effort with /reasoning.

  5. Check the key

    A list of models means the key works, and a 401 means SILK_API_KEY is empty or wrong.

    Shell
    curl -s https://api.silkcompute.com/models -H "Authorization: Bearer $SILK_API_KEY"

Other APIs that work with Hermes Agent

Hermes Agent's chat_completions transport takes any Chat Completions API, and three of the seven APIs for coding agents have a 1M-token context.

With a local model

Add a second provider whose api is a local Chat Completions server, and list one of the local models under models.

Silk Compute's own guide is at silkcompute.com/docs/hermes.