🆕 Module Addition
Support the addition of a new module to the Phare LLM benchmark, focused on evaluating a new task or category of AI Safety / Security risk
Includes dataset curation, expert validation, and domain-specific safety metrics.
Please let us know which task and specific LLM Safety / Security risk you'd like to see included in the Phare benchmark by email to [email protected].
Please let us know which task and specific LLM Safety / Security risk you'd like to see included in the Phare benchmark by email to [email protected].