Integrate third-party models with Hermes Agent

Bailian Token Plan Team Edition

1. Password-free logon to the terminal: Log on to the Simple Application Server console, find the target server card, click Remote Connection, and in the Connect with One-click on Workbench area, click Log on Now.

2. Subscribe to the Token Plan Team Edition plan and obtain an API key.

3. Configure access credentials: Run the following commands in the terminal. Replace YOUR_API_KEY with the API key for your Bailian Token Plan Team Edition.

   hermes config set model.provider custom
   hermes config set model.base_url https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1
   hermes config set model.api_key YOUR_API_KEY
   hermes config set model.default qwen3.6-plus

   These commands write the configuration to the ~/.hermes/config.yaml file. You can also edit the file directly and add the following content:

   model:
     default: qwen3.6-plus
     provider: custom
     base_url: https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1
     api_key: YOUR_API_KEY

4. Verify the configuration: After the configuration is complete, run the following command to send a test message:

   hermes chat -q "Hello"

   A valid response confirms a successful configuration.

5. Switch models: Use the -m parameter to specify a different model:

   hermes chat -m deepseek-v4-pro

Models supported by Bailian Token Plan Team Edition:

MiniMax Token Plan

1. Password-free logon to the terminal: Log on to the Simple Application Server console, find the target server card, click Remote Connection, and in the Connect with One-click on Workbench area, click Log on Now.

2. Subscribe to the MiniMax Token Plan and obtain an API key.

3. Configure the model: Run the following command in the terminal to start the model configuration wizard.

   hermes model

   In the interactive configuration, make the following selections:

   Select model provider  → MiniMax China (Chinese mainland endpoint)
   Enter API key          → Paste the MiniMax Token Plan API key you obtained in the previous step.
   default model          → MiniMax-M2.7

   Note

   For users outside of the Chinese mainland, select the MiniMax Global endpoint.

4. Verify the configuration: After the configuration is complete, run the following command to check the connection status.

   hermes doctor

5. Start Hermes Agent: After successful verification, run the following command in the terminal to start Hermes Agent.

   hermes

VolcEngine Ark Agent Plan

1. Password-free logon to the terminal: Log on to the Simple Application Server console, find the target server card, click Remote Connection, and in the Connect with One-click on Workbench area, click Log on Now.

2. Subscribe to the VolcEngine Ark Agent Plan and obtain an API key from the VolcEngine console.

3. Configure access credentials: Run the following commands in the terminal. Replace YOUR_ARK_API_KEY with the API key you obtained in the previous step.

   Warning

   Do not use https://ark.cn-beijing.volces.com/api/v3 as the base URL. Requests sent to this address do not consume your Agent Plan quota and will incur additional charges.

   hermes config set model.provider custom
   hermes config set model.base_url https://ark.cn-beijing.volces.com/api/plan/v3
   hermes config set model.api_key YOUR_ARK_API_KEY
   hermes config set model.default doubao-seed-2.0-pro

   These commands write the configuration to the ~/.hermes/config.yaml file. You can also edit the file directly and add the following content:

   model:
     default: doubao-seed-2.0-pro
     provider: custom
     base_url: https://ark.cn-beijing.volces.com/api/plan/v3
     api_key: YOUR_ARK_API_KEY

4. Verify the configuration: After the configuration is complete, run the following command to send a test message:

   hermes chat -q "Hello"

   A valid response confirms a successful configuration.

5. Switch models: To switch models, specify the model with the -m parameter:

   hermes chat -m glm-5.1

Models supported by the Agent Plan:

Tencent Token Plan

1. Password-free logon to the terminal: Log on to the Simple Application Server console, find the target server card, click Remote Connection, and in the Connect with One-click on Workbench area, click Log on Now.

2. Subscribe to the Tencent Cloud Token Plan and generate an API key (in the format sk-tp-xxx) in the TokenHub console.

3. Configure access credentials: Run the following commands in the terminal. Replace YOUR_API_KEY with the API key you obtained in the previous step.

   Note

   Model IDs must be configured in all lowercase.

   hermes config set model.provider custom
   hermes config set model.base_url https://api.lkeap.cloud.tencent.com/plan/v3
   hermes config set model.api_key YOUR_API_KEY
   hermes config set model.default hunyuan-2.0-thinking

   These commands write the configuration to the ~/.hermes/config.yaml file. You can also edit the file directly and add the following content:

   model:
     default: hunyuan-2.0-thinking
     provider: custom
     base_url: https://api.lkeap.cloud.tencent.com/plan/v3
     api_key: YOUR_API_KEY

4. Verify the configuration: After the configuration is complete, run the following command to send a test message:

   hermes chat -q "Hello"

   A valid response confirms a successful configuration.

5. Switch models: To switch models, specify the model with the -m parameter:

   hermes chat -m kimi-k2.5

Models supported by the Token Plan: