DeepSeek V4 Flash Vision is DeepSeek's language model. A vision-capable large language model optimized for fast multimodal reasoning with maximum effort chain-of-thought processing.
Specifications
Canonical IDdeepseek-v4-flash-vision
TypeLanguage
StatusActive
CreatorDeepSeekDeepSeek
Input ModalitiesImageText
Output ModalitiesText
Benchmarks
Intelligence Index
#41
Coding Index
#34
GPQA
#29
HLE
#59
Time to First Token
#420
SciCode
#56
LCR
#26
Output TPS
#15

Capabilities

Input2/5
Text
Image
Audio·
Video·
PDF·
Output1/5
Text
Image·
Audio·
Video·
Embedding·
Capabilities0/13
Reasoning·
Adaptive Reasoning·
Function Calling·
Parallel Function Calling·
Structured Outputs·
Native JSON Schema·
Web Search·
URL Context·
Computer Use·
Code Execution·
File Search·
Prompt Caching·
Assistant Prefill·

Versions

VersionReleasedContextInput / 1MOutput / 1MStatus
DeepSeek V4.731 Flash USAvailable
DeepSeek V4 420 FlashAvailable
DeepSeek V4.4.1 FlashAvailable
DeepSeek V4.1 Flash Expires Exp1.0M$0.220$0.660Available
DeepSeek V4.1 Flash Beta1.0MAvailable
DeepSeek V4 Flash Vision Exp1.0M$0.220$0.660Available
DeepSeek V4 Flash1.3M$0.030$0.177Deprecating
DeepSeek V4 Flash VisionCurrent
DeepSeek V4 Flash 7311.0M$0.140$0.280Available
DeepSeek V4 Flash Thinking200K$0.250$1.75Available
DeepSeek V4 Flash USAvailable

Other Models

ModelTierReleasedContextInput / 1MOutput / 1M
DeepSeek V4 ProPro1.0M$0.624$1.98
DeepSeek V4 ProPro
DeepSeek V4 ProPro1.0M$0.435$0.870
DeepSeek V4 Pro 813Pro1.0M$1.32$3.96
DeepSeek V4 Pro USPro

Model IDs

deepseek-v4-flash-vision