files api

1 article

← All topics

DeepSeek ships V4-Flash-Vision-Exp — its first multimodal model, matching V4-Flash on text, adding image input
deepseekdeepseek-v4-flashvision-expmultimodalimage-input+2

DeepSeek's experimental vision model matches V4-Flash on text tasks and adds image input at the same pricing; new Files API lets you upload once and reuse across requests.