DeepSeek-v4-flash-vision-exp

DeepSeek's deepseek-v4-flash-vision-exp model is a multimodal AI that accepts images alongside text to perform tasks like image description, text extraction from screenshots, and chart analysis. The model supports JPEG, PNG, GIF, and WebP formats and offers three methods for providing images: base64-encoded inline images, external URLs, or references to files uploaded via the Files API, each with different size and reusability constraints.