If you discover a security vulnerability in Granite Switch, please report it responsibly.
Do not open a public GitHub issue for security vulnerabilities.
Instead, please report security issues by emailing the maintainers directly or through GitHub's private vulnerability reporting feature.
Security updates are provided for the latest release only.
When using Granite Switch:
- Model Access: Ensure proper access controls when serving models via vLLM or other inference servers
- API Security: When deploying with vLLM's OpenAI-compatible API, implement appropriate authentication
- Dependencies: Keep dependencies updated to receive security patches
We appreciate your efforts to responsibly disclose your findings and will make every effort to acknowledge your contributions.