DeepSeek V4 Flash Vision Exp
Retired Sep 2026An experimental multimodal variant of DeepSeek V4 Flash that adds image input over the same text capabilities in agents, reasoning and world knowledge.
- Context
- 1M tokens
- Max output
- 384K
- Released
Capabilities
What DeepSeek V4 Flash Vision Exp accepts and can do, as published by DeepSeek.
- Text Accepts and generates natural-language text.
- Image Accepts images as input alongside text.
- Reasoning Thinks before it answers, always on or as a switchable mode.
- Function calling Connect to external tools, APIs, and systems.
- Structured output Return responses in structured formats like JSON.
- Web search Search the web for up-to-date information.
Common questions
What is DeepSeek V4 Flash Vision Exp good for?
Agent work that needs image input alongside text. DeepSeek says it matches V4 Flash on text tasks.
When is DeepSeek V4 Flash Vision Exp not a good fit?
New work: DeepSeek retired it and now routes the model name to V4.1 Flash, which takes images natively. The weights were never published, so it cannot be self-hosted.
Can I self-host DeepSeek V4 Flash Vision Exp?
No. DeepSeek does not publish the weights; DeepSeek V4 Flash Vision Exp is available only through hosted APIs.
More from DeepSeek
-
DeepSeek V4 Flash
Open weights
1M context
From $0.14 / $0.28 /1M in / out at the cheapest provider
-
DeepSeek V4 Pro
Open weights
1M context
From $0.91 / $2.72 /1M in / out at the cheapest provider
-
DeepSeek V4.1 Flash
Open weights
1M context
From $0.035 / $0.50 /1M in / out at the cheapest provider
-
DeepSeek Prover V2
Open weights
164K context
No hosted price
-
DeepSeek R1 Distill Llama 70B
Open weights
131K context
No hosted price
-
DeepSeek V3.1
Open weights
131K context
From $0.672 / $2.016 /1M in / out at the cheapest provider
-
DeepSeek V3.1 Terminus
Open weights
131K context
No hosted price
-
DeepSeek V3.2
Open weights
131K context
From $0.50 / $1.60 /1M in / out at the cheapest provider