Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
EVA‑CLIP‑18Bhuggingface.co · Low evidenceEVA-CLIP-18B is an open-source contrastive language-image pretraining model with 18 billion parameters, achieving 80.7% zero-shot top-1 accuracy across 27 image classification benchmarks. It demonstrates consistent performance improvements through model scaling while using a publicly available training dataset of 2 billion image-text pairs.
Parse Score