DeepSeek Vision Model Officially Launches on API: Pricing Identical to V4 Flash

DeepSeek has officially launched its new vision model, deepseek-v4-flash-vision-exp. The official API documentation now lists it alongside V4 Flash and V4 Pro, allowing developers to directly pass images via the DeepSeek API. The model supports a 1 million Token context window, with a maximum output of 384,000 Tokens, and also supports JSON Output, Tool Calls, Responses API, and the Anthropic API. Pricing is directly aligned with V4 Flash. For every million Tokens of input, when the cache is missed, the peak period costs 3 yuan and the off-peak period costs 1.5 yuan; with cache hits, it costs just 0.1 yuan and 0.05 yuan, respectively. For every million Tokens of output, the peak period costs 9 yuan and the off-peak period costs 4.5 yuan. Compared to V4 Pro, the same tier is only one-third the price. Images are not charged separately per image. DeepSeek converts images into Tokens based on their dimensions and then bills them together with text Tokens. The peak period is Beijing time 9:00–12:00 and 14:00–18:00, with prices halved during all other times.