← Library · Frontier

EleutherAI Releases Largest Open Source Vision-Language Model, 'CLIP-ViT/L'

EleutherAI, a collective of independent AI researchers, has released 'CLIP-ViT/L-14', a large open-source vision-language model trained on a massive dataset. This model expands upon OpenAI's CLIP architecture, offering enhanced capabilities for tasks requiring understanding both images and text, such as image retrieval and zero-shot classification.

Why it matters

By providing a larger, open-source vision-language model, EleutherAI accelerates research and application development in multimodal AI. It allows broader experimentation and fine-tuning for various computer vision tasks.

Learn one new AI thing every day.

Daily Deck sends you seven plain-English cards like this every morning. Free.

Start free