Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
Long-CLIPgithub.com · Low evidenceLong CLIP is an open-source extension of OpenAI's CLIP model that increases its maximum text input length from 77 to 248 tokens. It improves long-caption text-image retrieval by 20% and traditional retrieval by 6%, and can be directly integrated into any system requiring long text understanding.
Parse Score