Nexent Doc

English

简体中文

English

简体中文

Appearance

Nexent FAQ ​

This FAQ addresses common questions and issues you might encounter while installing and using Nexent. For the basic installation steps, please refer to the Installation & Deployment. For basic using instructions, please refer to the User Guide.

🚫 Common Errors & Operations ​

🌐 Network Connection Issues ​

  • Q: How can a Docker container access models deployed on the host machine (e.g., Ollama)?

    • A: Since localhost inside the container refers to the container itself, use one of these methods to connect to host services:

      Option 1: Use Docker's special DNS name host.docker.internal

      Supported environments: Mac/Windows and newer Docker Desktop versions (Linux version also supported)

      bash
      http://host.docker.internal:11434/v1

      Option 2: Use host machine's actual IP (ensure firewall allows access)

      bash
      http://[HOST_IP]:11434/v1

      Option 3: Modify Docker Compose configuration

      Add to your docker-compose.yaml file:

      yaml
      extra_hosts:
  - "host.docker.internal:host-gateway"

🔌 Port Conflicts ​

  • Q: Port 3000 is already in use. How can I change it?
    • A: You can modify the port in the Docker Compose configuration file.

📦 Container Issues ​

  • Q: How do I check container logs?
    • A: Use docker logs <container_name> to view logs for specific containers.

🔍 Troubleshooting ​

🔢 Model Connection Issues ​

  • Q: Why can't my model connect?

    • A: Check the following:

      1. Correct API endpoint: Ensure you're using the right base URL
      2. Valid API key: Verify your API key has proper permissions
      3. Model name: Confirm the model identifier is correct
      4. Network access: Ensure your deployment can reach the provider's servers

      For model setup instruction, see Model Management in User Guide.

  • Q: Multi-turn chats fail when using the official DeepSeek API. How can I resolve this?

    • A: The official DeepSeek API only accepts text payloads, but Nexent sends multimodal payloads, so multi-turn calls are rejected. Use a provider such as SiliconFlow that exposes DeepSeek models with multimodal compatibility. Our requests look like:
    python
    { "role":"user", "content":[ { "type":"text", "text":"prompt" } ] }

    whereas DeepSeek expects:

    python
    { "role":"user", "content":"prompt" }

🐛 Known Issues & Feedback ​

If you encounter any issues or want to check the latest status of known issues, please visit:

  • Search similar issues: GitHub Issues - Search here to see if a similar issue has already been reported
  • Discuss issues: GitHub Discussions - Discuss problems and solutions with the community here

💡 Need Help ​

If your question isn't answered here: