> ## Documentation Index
> Fetch the complete documentation index at: https://docs.bubench.lexmount.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Get Help

> Having trouble? Here's how to get help

If you encounter issues while using browseruse-bench, here are ways to get help.

## Common Issues

### Installation Issues

<AccordionGroup>
  <Accordion title="pip install failed">
    Try installing with uv:

    ```bash theme={null}
    pip install uv
    uv sync
    ```
  </Accordion>

  <Accordion title="Chrome browser not found">
    Ensure Chrome is installed and in PATH, or use Lexmount Cloud Browser.
  </Accordion>

  <Accordion title="Agent-TARS installation failed">
    Ensure Node.js version >= 18:

    ```bash theme={null}
    node --version
    npm install -g @agent-tars/cli@0.3.0
    ```
  </Accordion>
</AccordionGroup>

### Runtime Issues

<AccordionGroup>
  <Accordion title="Task Timeout">
    Increase timeout duration:

    ```bash theme={null}
    bubench run --agent browser-use --data LexBench-Browser --timeout 600
    ```
  </Accordion>

  <Accordion title="Invalid API Key">
    Check if API Key in `.env` file is correct:

    ```bash theme={null}
    cat .env | grep API_KEY
    ```
  </Accordion>

  <Accordion title="Cloud Browser Connection Failed">
    1. Check `LEXMOUNT_API_KEY` and `LEXMOUNT_PROJECT_ID`
    2. Confirm network access to Lexmount API
  </Accordion>
</AccordionGroup>

### Evaluation Issues

<AccordionGroup>
  <Accordion title="Evaluation Model Call Failed">
    Check `OPENAI_API_KEY` and the `eval.model` / `eval.base_url` section in `config.yaml`.
  </Accordion>

  <Accordion title="All Evaluation Results Failed">
    Try lowering the score threshold:

    ```bash theme={null}
    bubench eval --agent browser-use --data LexBench-Browser --model-id bu-2-0 --score-threshold 50
    ```
  </Accordion>
</AccordionGroup>

## View Logs

### Runtime Logs

Runtime logs are saved in the `output/logs/` directory:

```bash theme={null}
# View latest logs
ls -lt output/logs/ | head -5

# Tail log content
tail -f output/logs/<latest_log>.log
```

## Submit Issue

If none of the above methods solve your problem, please submit an Issue on GitHub:

1. Visit [GitHub Issues](https://github.com/lexmount/browseruse-bench/issues)
2. Click "New Issue"
3. Provide the following information:
   * Problem description
   * Reproduction steps
   * Error logs
   * System environment (OS, Python version, etc.)

### Issue Template

```markdown theme={null}
## Problem Description
Briefly describe the problem encountered

## Reproduction Steps
1. Run command `...`
2. See error `...`

## Error Logs
```

Paste error logs

```

## Environment Info
- OS: macOS / Linux / Windows
- Python: 3.10 / 3.11 / 3.12
- browseruse-bench version: ...
```

## Community Support

* **GitHub Discussions**: Questions and discussions
* **GitHub Issues**: Report bugs and feature requests

## Commercial Support

For commercial support or enterprise deployment inquiries, please contact [support@lexmount.com](mailto:support@lexmount.com).