DeepSeek-V4-Flash-Vision-Exp

An experimental V4 Flash API model adding image input.

CurrentReasoningCodingAgentMultimodalWeights not released

MODEL NUMBER068

Specifications

Only published specifications are shown.

Total parameters

284B

Active parameters

13B

Architecture

MoE

Context

Input modalities

Text · Image

Open weights

Status

Current

An experimental V4 Flash API model adding image input.

PreviousDeepSeek-V4-Flash-0731284B · 1M ctx

CurrentDeepSeek-V4-Flash-Vision-Exp284B · 1M ctx

New capabilities Vision

DSA

VisionAdjustable reasoningAgentResponses API

2026-08-21Preview

DeepSeek-V4-Flash-Vision-Exp launched in the multimodal API.

Select a card to view the related model.