Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
LlavaGuardgithub.com · Low evidenceLlavaGuard is a VLM-based framework for evaluating the safety compliance of visual content, designed for dataset curation and safeguarding generative models. It provides open pre-trained weights, a customizable safety taxonomy, and tools for direct inference, fine-tuning, and full model training.
Parse Score