DeepSeek V4 Flash Vision Exp (batch)
Archived
DeepSeek🇨🇳
deepseek/deepseek-v4-flash-vision-exp:batchBatch-processing variant of DeepSeek V4 Flash Vision Exp, the experimental multimodal version of DeepSeek V4 Flash (284B total, 13B active parameters, 1M-token context) that accepts image input for document and chart understanding, visual question answering, and multimodal agent workflows. Same capabilities as the standard model, offered at lower cost via asynchronous batch processing.
Cost rate
0.1x
Context
1M
Released
Aug 21, 2026
Input
TextImage
Output
Text
Support
ReasoningTool callingStructured outputs