DeepSeek V4 420 Flash is DeepSeek's language model. A lightweight, low-latency variant of DeepSeek's V4 large language model tuned for fast reasoning responses.
Specifications
Canonical IDdeepseek-v4-420-flash
TypeLanguage
StatusActive
CreatorDeepSeekDeepSeek
Input ModalitiesText
Output ModalitiesText
Benchmarks
Intelligence Index
#45
Coding Index
#43
GPQA
#47
HLE
#47
IFBench
#10
Time to First Token
#111
SciCode
#68
LCR
#88
TerminalBench Hard
#64
TAU2
#28
Output TPS
#344

Capabilities

Input1/5
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 420 FlashCurrent
DeepSeek V4 Flash Vision Exp1.0M$0.220$0.660Available
DeepSeek V4 Flash1.3M$0.040$0.080Available
DeepSeek V4 Flash Thinking200K$0.250$1.75Available

Other Models

ModelTierReleasedContextInput / 1MOutput / 1M
DeepSeek V4 ProPro1.0M$1.12$3.37
DeepSeek V4 ProPro
DeepSeek V4 ProPro1.0M$0.435$0.870

Model IDs

deepseek-v4-420-flash
deepseek-v4-flash-0420
deepseek-v4-flash-0420-high