DeepSeek-V4-Flash-Vision-Exp
DeepSeek-V4-Flash-Vision-Exp is DeepSeek's first experimental multimodal model in the V4 family. The official Hugging Face repository describes a vision-augmented DeepSeek-V4-Flash checkpoint with MIT-licensed weights, about 305B parameters, text and image input, and reference serving recipes for vLLM and SGLang.
Newer version: DeepSeek-V4.1-Flash
Provider
DeepSeek
Model family
DeepSeek V4
Experimental multimodal MoE LLM
Cost tier
Flash Vision Exp
Status
Deprecated
Release Aug 31, 2026
Why teams choose it
DeepSeek positions this as an experimental vision add-on to V4-Flash
not a replacement for text-only DeepSeek-V4-Flash.
The Hugging Face card reports comparable text-agent scores to V4-Flash-0731 with
stronger multimodal agent results.
Official recipes document vLLM and SGLang serving
including DSpark speculative decoding from the same checkpoint.
Tradeoffs to know
- The -Exp suffix means experimental; APIs, weights, and quality can change before a final V4 vision release.
- A ~305B checkpoint is a multi-GPU download and serve; most teams will still use the hosted API for trials.
- Benchmark tables are vendor-reported and use DeepSeek's harness settings; reproduce before substituting it for V4-Flash.
Technical specs
- Inputs
- text, image
- Outputs
- text
- Capabilities
- multimodal reasoning, image understanding, coding agents, tool use, open-weight serving
- License
- MIT
- Source
- https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-Vision-Exp
- Deep Swe
- 59.3
- Parameters
- 305B
Benchmarks
- Terminal Bench 2 1
- 83.9
- Toolathlon Verified
- 75.9
Related models
DeepSeek
DeepSeek Coder V2
Catalog entry for this named release; see the provider’s official documentation for modalities, pricing, and context…
DeepSeek
DeepSeek-V4.1-Flash
DeepSeek-V4.1-Flash is DeepSeek's September 10, 2026 multimodal Mixture-of-Experts model and the smallest member of i…
DeepSeek
DeepSeek-V3.2
DeepSeek's documented successor to the V3.2 experimental line, positioned in official DeepSeek API news as live on ap…
DeepSeek-V4-Flash-Vision-Exp FAQ
What is DeepSeek-V4-Flash-Vision-Exp?
DeepSeek-V4-Flash-Vision-Exp is DeepSeek's first experimental multimodal model in the V4 family. The official Hugging Face repository describes a vision-augmented DeepSeek-V4-Flash checkpoint with MIT-licensed weights, about 305B parameters, text and image input, and reference serving recipes for vLLM and SGLang.
When does DeepSeek-V4-Flash-Vision-Exp fit best?
Image-grounded coding and tool agents
What should teams watch out for with DeepSeek-V4-Flash-Vision-Exp?
The -Exp suffix means experimental; APIs, weights, and quality can change before a final V4 vision release.
Explore next
Models, tools, and comparisons that connect to this reference.