---
title: "Deepseek's new vision model approaches Opus 4.8 on agent tasks"
url: https://www.parallelquant.com/posts/deepseek-s-new-vision-model-approaches-opus-4-8-on-agent-tasks-396a74
source_name: "The Decoder"
source_url: https://the-decoder.com/deepseek-releases-experimental-flash-vision-model-that-rivals-opus-4-8-on-agent-benchmarks/
published: 2026-08-21T19:08:27.000Z
topics: []
publisher: "Parallel Quant"
---

# Deepseek's new vision model approaches Opus 4.8 on agent tasks

*2026-08-21 · Source: [The Decoder](https://the-decoder.com/deepseek-releases-experimental-flash-vision-model-that-rivals-opus-4-8-on-agent-benchmarks/)*

Deepseek released V4-Flash-Vision-Exp, an experimental multimodal model that adds image understanding to its V4-Flash text model. On Deepseek's own multimodal agent benchmarks, the model approaches Anthropic's Opus 4.8 and sometimes surpasses it.

**Why it matters:** A Chinese lab closing the gap with a leading US frontier model on agent benchmarks, even using self-reported figures, adds another data point to the narrative that the US-China AI capability gap is narrowing rather than widening.

---
Read the original: https://the-decoder.com/deepseek-releases-experimental-flash-vision-model-that-rivals-opus-4-8-on-agent-benchmarks/
Canonical: https://www.parallelquant.com/posts/deepseek-s-new-vision-model-approaches-opus-4-8-on-agent-tasks-396a74
Published by Parallel Quant — https://www.parallelquant.com