DeepSeek·DeepSeek V / R

DeepSeek-V4-Flash-Preview

The efficient V4 preview with 284B total, 13B active parameters, and a 1M context window.

LegacyLanguageReasoningCodingAgentOpen weights

MODEL NUMBER102

First public date: 2026-04-24
Organization: DeepSeek
Family: DeepSeek V / R

Specifications

Only published specifications are shown.

Total parameters: 284B
Active parameters: 13B
Architecture: MoE
Context: 1,000,000 tokens
Input modalities: Text
Open weights: Yes
License: DeepSeek Model License
Status: Legacy

Overview

Release dates

April 24, 2026 — DeepSeek released public previews and open weights for V4 Pro and Flash together. ↗

Architecture & capabilities

Attention

DSA

Model capabilities

Hybrid reasoningAgentCodingLong context

Release history

2026-04-24Preview

DeepSeek released public previews and open weights for V4 Pro and Flash together.

View source

Related models

Select a card to view the related model.

SiblingDeepSeek-V4-Pro-PreviewView details →SuccessorDeepSeek-V4-Flash-0731View details →

Sources

DeepSeek-V4 API releaseDeepSeek

Official · Verified

Suggest a correction

Copy this template, add the field, proposed value, and primary source, and share it with the archive maintainer.

Model ID: deepseek-v4-flash-preview
Page: https://llmlineage.com/en/models/deepseek-v4-flash-preview
Field:
Proposed value:
Primary source URL:
Reason: