Skip to main content

​
Troubleshooting

This guide covers common issues you might encounter with Jan and how to resolve them.

​
Getting Error Logs

Error logs are essential for troubleshooting. You can access them in two ways:

​
Through Jan Interface

  1. Open System Monitor in the footer
  2. Choose App Log

​
Through Terminal

Application Logs: 
# macOS/Linux
tail -n 50 ~/Library/Application\ Support/Jan/data/logs/app.log

# Linux (alternative path)
tail -n 50 ~/.config/Jan/logs/app.log
Server Logs: 
tail -n 50 ~/Library/Application\ Support/Jan/data/logs/cortex.log
Redact any private or sensitive information when sharing logs. We retain logs for only 24 hours.

​
Common Issues

​
Broken Build

If Jan is stuck in a broken build after installation:
  1. Uninstall Jan
  2. Delete Application Data:
rm -rf ~/Library/Application\ Support/Jan
  1. For versions before 0.4.2, kill background processes:
ps aux | grep nitro
kill -9 <PID>
  1. Download the latest version
Ensure Jan is completely removed from all user accounts on shared devices before reinstalling.

​
NVIDIA GPU Not Working

If Jan doesn’t utilize your NVIDIA GPU:

​
1. Verify Hardware Detection

Windows:
  • Right-click desktop → NVIDIA Control Panel
  • Or check Device Manager → Display Adapters
Linux: 
lspci | grep -i nvidia

​
2. Install Required Components

NVIDIA Driver (470.63.01 or higher):
  1. Install NVIDIA Driver
  2. Verify: nvidia-smi
CUDA Toolkit (11.7 or higher):
  1. Install CUDA toolkit
  2. Verify: nvcc --version
Linux Additional Setup: 
sudo apt update
sudo apt install gcc-11 g++-11 cpp-11
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda/lib64

​
3. Enable GPU Acceleration

  1. Open Settings > Hardware
  2. Verify GPU Acceleration is enabled
  3. Check selected GPU(s) in System Monitor (footer)

​
4. Check GPU Settings

  1. Go to Settings > General > Data Folder
  2. Click Open Containing Folder
  3. Open settings.json
Verify configuration: 
{
  "run_mode": "gpu",
  "nvidia_driver": {
    "exist": true,
    "version": "531.18"
  },
  "cuda": {
    "exist": true,
    "version": "12"
  },
  "gpus": [
    {
      "id": "0",
      "vram": "12282"
    }
  ]
}

​
“Failed to Fetch” or “Something’s Amiss” Errors

1. Check System Requirements
  • Verify your device meets minimum requirements
  • RAM: Choose models using less than 80% of available RAM
    • 8GB systems: Models under 6GB
    • 16GB systems: Models under 13GB
2. Adjust Model Parameters
  • Check ngl (GPU layers) setting in Engine Settings
  • Start with lower values and increase gradually
3. Check Port Conflicts If logs show “Bind address failed”: 
# macOS
netstat -an | grep 39291

# Windows
netstat -ano | find "39291"
tasklist /fi "PID eq 39291"

# Linux
netstat -anpe | grep "39291"
Default Jan ports:
  • API Server: 1337
  • Documentation: 3001
4. Factory Reset
  1. Go to Settings > Advanced Settings
  2. Click Reset under Reset To Factory Settings
This deletes all chat history, models, and settings.
5. Clean Installation

​
Permission Denied Error

If you see permission errors during installation: 
sudo chown -R $(whoami) ~/.npm

​
OpenAI “Unexpected Token” Error

1. API Key Setup 2. Regional Access
  • Use a VPN if in a restricted region
  • Verify network access to OpenAI endpoints

​
Need More Help?

  1. Copy your error logs
  2. Visit our Discord
  3. Post in #🆘|jan-help channel
Check logs before sharing. We retain logs for only 24 hours.

Build docs developers (and LLMs) love

Get started for freeTalk to us