What is DeepSeek V4 Flash 0731?

DeepSeek V4 Flash 0731 is cataloged under DeepSeek / DeepSeek with canonical ID deepseek/deepseek-v4-flash-0731.

DEDeepSeek/ DeepSeek

DeepSeek V4 Flash 0731

deepseek/deepseek-v4-flash-0731

DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total. This re-post-trained revision is suited for coding, reasoning, and agent workflows.

Compare

Model information

Released

2026-07-31

Context window

Max output

66K

Model parameters

304B

Weights

Open

Modalities

Input

Text

Output

Text

Identity

Version: deepseek-v4-flash-0731
License: MIT

Resources

Hugging Face

Capabilities

Open weightReasoningStructuredTemperatureTool call

Providers

Provider	Provider model ID	Protocol	Context	Max output	Capabilities	Notes	Pricing	Streaming
OpenRouter	deepseek/deepseek-v4-flash-0731	openai_chat	1M	66K	ReasoningTool callStructuredTemperature	Reasoning levels: max, high, low, Default reasoning: high, Hugging Face: deepseek-ai/DeepSeek-V4-Flash-0731	input $0.09/M · output $0.18/M cache read $0.018/M	Yes
NanoGPT	deepseek/deepseek-v4-flash-0731	openai_chat	1M	66K	ReasoningTool callStructuredTemperature	-	input $0.14/M · output $0.28/M cache read $0.014/M	Yes
TensorX	deepseek/deepseek-v4-flash-0731	openai_chat	1M	66K	ReasoningTool callStructuredTemperature	-	input $0.25/M · output $0.3/M cache read $0.06/M	Yes
Kilo Gateway	deepseek/deepseek-v4-flash-0731	openai_chat	1M	66K	ReasoningTool callStructuredTemperature	-	input $0.14/M · output $0.28/M cache read $0.028/M	Yes
Vercel AI Gateway	deepseek/deepseek-v4-flash-0731	openai_chat	1M	66K	ReasoningTool callStructuredTemperature	-	input $0.13/M · output $0.26/M cache read $0.028/M	Yes
Ambient	deepseek/deepseek-v4-flash-0731	openai_chat	1M	66K	ReasoningTool callStructuredTemperature	-	input $0.14/M · output $0.28/M cache read $0.028/M · cache write $0/M	Yes
DeepSeek	deepseek-v4-flash-0731	custom	1M	66K	-	-	-	Yes