Troubleshooting

Broken Build

To resolve the issue where your Jan is stuck in a broken build after installation.

By following these steps, you can cleanly uninstall and reinstall Jan, ensuring a smooth and error-free experience with the latest version.

Troubleshooting NVIDIA GPU

To resolve issues when the Jan app does not utilize the NVIDIA GPU on Windows and Linux systems.

1. Ensure GPU Mode Requirements

2. Switch to GPU Mode

If your system supports it, Jan defaults to CPU mode but automatically switches to GPU mode, selecting the GPU with the highest VRAM. Check this setting in Settings > Advanced Settings.

Troubleshooting Tips

If GPU mode isn't enabled by default:

1. Confirm that you have installed an NVIDIA driver supporting CUDA 11.7 or higher. Refer to CUDA compatibility (opens in a new tab).
2. Ensure compatibility of the CUDA toolkit with your NVIDIA driver. Refer to CUDA compatibility (opens in a new tab).
3. Add CUDA's .so libraries to the LD_LIBRARY_PATH for Linux. Ensure that CUDA's .dll libraries are in the PATH for Windows. Refer to Windows setup (opens in a new tab).

If you encounter an error message indicating that loading a model requires additional dependencies, follow these steps:

1. Click on Install Additional Dependencies in the error message.
2. You will be redirected to the Tensor RT LLM Inference Engine section.
3. Click the Install button to install additional dependencies.

3. Check GPU Settings

1. Navigate to Settings > Advanced Settings > Jan Data Folder to access GPU settings.
2. Open the settings.json file in the settings folder. Here's an example:

4. Restart Jan

Restart the Jan application to make sure it works.

Troubleshooting Tips

• Ensure the nvidia_driver and cuda fields indicate installed software.
• If gpus field is empty or lacks your GPU, check the NVIDIA driver and CUDA toolkit installations.
• For further assistance, share the settings.json file.

Tested Configurations

• Windows 11 Pro 64-bit:

  • GPU: NVIDIA GeForce RTX 4070ti
  • CUDA: 12.2
  • NVIDIA driver: 531.18 (Bare metal)

• Ubuntu 22.04 LTS:

  • GPU: NVIDIA GeForce RTX 4070ti
  • CUDA: 12.2
  • NVIDIA driver: 545 (Bare metal)

• Ubuntu 20.04 LTS:

  • GPU: NVIDIA GeForce GTX 1660ti
  • CUDA: 12.1
  • NVIDIA driver: 535 (Proxmox VM passthrough GPU)

• Ubuntu 18.04 LTS:

  • GPU: NVIDIA GeForce GTX 1660ti
  • CUDA: 12.1
  • NVIDIA driver: 535 (Proxmox VM passthrough GPU)

Common Issues and Solutions

1. If the issue persists, install the Nightly version.
2. Ensure your (V)RAM is accessible; some users with virtual RAM may require additional configuration.
3. Seek assistance in Jan Discord (opens in a new tab).

How to Get Error Logs

To get the error logs of your Jan application, follow the steps below:

Jan Application

1. Navigate to the main dashboard.
2. Click the gear icon (⚙️) on the bottom left of your screen.
3. Under the Settings screen, click the Advanced Settings.
4. On the Jan Data Folder click the folder icon (📂) to access the data.
5. Click the logs folder.

Jan UI

1. Open your Unix or Linux terminal.
2. Use the following commands to get the recent 50 lines of log files:

Jan API Server

1. Open your Unix or Linux terminal.
2. Use the following commands to get the recent 50 lines of log files:

Permission Denied

When running Jan, you might encounter the following error message:

Permission problems mainly cause this error during installation. To resolve this issue, follow these steps:

1. Open your terminal.

2. Execute the following command to change ownership of the ~/.npm directory to the current user:

Something's Amiss

When you start a chat with a model and encounter a Something's Amiss error, here's how to resolve it:

1. Ensure your OS is up to date.
2. Choose a model smaller than 80% of your hardware's V/RAM. For example, on an 8GB machine, opt for models smaller than 6 GB.
3. Install the latest Nightly release or clear the application cache when reinstalling Jan.
4. Confirm your V/RAM accessibility, mainly if using virtual RAM.
5. Nvidia GPU users should download CUDA (opens in a new tab).
6. Linux users, ensure your system meets the requirements of gcc 11, g++ 11, cpp 11, or higher. Refer to this link for details.
7. You might use the wrong port when you check the app logs and encounter the Bind address failed at 127.0.0.1:3928 error. To check the port status, try using the netstat command, like the following:

Undefined Issue

Encountering an undefined issue in Jan is caused by errors related to the Nitro tool or other internal processes. It can be resolved through the following steps:

1. Clear the Jan folder and then reopen the application to determine if the problem persists.
2. Manually run the nitro tool located at ~/jan/extensions/@janhq/inference-nitro-extensions/dist/bin/(your-os)/nitro to check for error messages.
3. Address any identified nitro error messages and reassess the issue's persistence.
4. Reopen Jan to determine if the problem has been resolved after addressing any identified errors.
5. If the issue persists, please share the app logs via Jan Discord (opens in a new tab) for further assistance and troubleshooting.

Unexpected Token

Encountering the Unexpected token error when initiating a chat with OpenAI models is mainly caused by your OpenAI key or where you access your OpenAI from. This issue can be solved through the following steps:

1. Obtain an OpenAI API key from OpenAI's developer platform (opens in a new tab) and integrate it into your application.

2. Using a VPN could potentially solve the issue, especially if it's related to region locking for accessing OpenAI services. Connecting through a VPN may bypass such restrictions and successfully initiate chats with OpenAI models.