DeepSeek-V4-Flash-Vision-Exp
DeepSeek-V4-Flash-Vision-Exp is an experimental multimodal API model (deepseek-v4-flash-vision-exp) that matches V4-Flash on text, agents, and reasoning while boosting multimodal-agent benchmarks to near Opus-4.8. It supports Chat Completions, Messages, and Responses APIs with images billed at V4-Flash rates (up to 384 tokens each) via base64, URL, or the Files API, which launched the same day with free uploads and reuse by file_id; DeepSeek Harness 0.1.1 includes day-one support.