DeepSeek-V4-Flash-Vision-Exp has been launched on the DeepSeek API platform. According to news on August 21, the multi-modal visual understanding model DeepSeek-V4-Flash-Vision-Exp has been launched on the DeepSeek API platform, supporting Chat Compl

DeepSeek-V4-Flash-Vision-Exp is launched on the DeepSeek API platform. According to news on August 21, the multi-modal visual understanding model DeepSeek-V4-Flash-Vision-Exp has been launched on the DeepSeek API platform, supporting calls in three formats: Chat Completions, Messages, and Responses. In terms of image understanding, the DeepSeek-v4-flash-vision-exp model supports inputting images in addition to text, and allows the model to describe images, recognize text in screenshots, analyze charts, etc. In terms of price, the input cache hit for one million tokens starts at 0.05 yuan, the cache miss starts at 1.5 yuan, and the output for one million tokens starts at 4.5 yuan. via cnBeta.COM - Chinese industry information website (author: Source: Sanyan Technology)