AI Model Watch
Deprecations / Migrate

Migrate off DeepSeek-V4-Flash-Vision-Exp

Retired

DeepSeek-V4-Flash-Vision-Exp has been retired and shut down on 10 Sep 2026. DeepSeek's official replacement is DeepSeek-V4.1-Flash. The cheapest comparable model we track is Llama 4 Scout (17B-16E Instruct) (Meta) at $0.15 per 1M tokens blended.

Last verified: 26 Sep 2026 · sourced from official provider documentation

StatusRetiredDeepSeek
Shut down10 Sep 2026plan ahead
Official replacementDeepSeek-V4.1-Flash$0.525 /M blendedCheapest alternativeLlama 4 Scout (17B-16E Instruct)$0.15 /M blended
What changes

DeepSeek-V4-Flash-Vision-Exp → DeepSeek-V4.1-Flash

Moving to DeepSeek-V4.1-Flash means 1.3× cheaper ($0.66 → $0.525 /1M blended).

SpecDeepSeek-V4-Flash-Vision-ExpDeepSeek-V4.1-Flash
StatusRetiredGA
Blended price /1M$0.66$0.525
Input / output /1M$0.44 / $1.32$0.3 / $1.20
Context window1M1M
Max output384.000384.000
Modalitytext, imagetext, image
Knowledge cutoff——

API string was deepseek-v4-flash-vision-exp → switch to deepseek-flash. See DeepSeek-V4.1-Flash's full spec sheet →

Cheapest alternatives

Budget-friendly replacements for DeepSeek-V4-Flash-Vision-Exp

Generally-available models that match DeepSeek-V4-Flash-Vision-Exp's capabilities (and at least half its 1M context), ranked by blended price. Every number is computed from official catalog prices.

ModelBlended /1MInput / outputContext
Llama 4 Scout (17B-16E Instruct)
Meta
$0.15$0.1 / $0.310M
Qwen3.5-Flash
Alibaba
$0.175$0.1 / $0.41M
GPT-6 Luna
OpenAI
$0.2$0.1 / $0.51.1M
FAQ

Migrating off DeepSeek-V4-Flash-Vision-Exp

What replaces DeepSeek-V4-Flash-Vision-Exp?

DeepSeek's recommended migration for DeepSeek-V4-Flash-Vision-Exp is DeepSeek-V4.1-Flash. The cheapest comparable model in our catalog is Llama 4 Scout (17B-16E Instruct) (Meta) at $0.15/1M blended.

Is DeepSeek-V4-Flash-Vision-Exp still available?

No — DeepSeek-V4-Flash-Vision-Exp has been retired (shut down 10 Sep 2026) and is no longer available via DeepSeek's API.

What is the cheapest alternative to DeepSeek-V4-Flash-Vision-Exp?

Llama 4 Scout (17B-16E Instruct) (Meta) is the cheapest generally-available alternative we track that matches DeepSeek-V4-Flash-Vision-Exp's capabilities, at $0.1 per 1M input and $0.3 per 1M output tokens ($0.15/1M blended).

Browse the full deprecation watch → or see DeepSeek-V4-Flash-Vision-Exp's full spec sheet →