deepseek-ai/DeepSeek-V4-Flash-Vision-Exp — new model trending #10 on Hugging Face
DeepSeek released DeepSeek-V4-Flash-Vision-Exp, its first experimental multimodal model, with large multimodal agent benchmark gains over V4-Flash-0731.
DeepSeek AI published DeepSeek-V4-Flash-Vision-Exp, an experimental multimodal model built on the DeepSeek-V4-Flash architecture with added visual modules and continued training. It scores 83.9 on Terminal Bench 2.1 and 36.5 on ApexBench Pass@1 versus 26.2 for DeepSeek-V4-Flash-0731, while remaining comparable to Opus-4.8 on several benchmarks. The MIT-licensed repository ships a tokenizer, OpenAI-style and TXT prompt encoding, and a minimal PyTorch inference implementation, with vLLM and SGLang deployment recipes.
- First experimental multimodal model in the DeepSeek-V4 family
- Improves ApexBench Pass@1 to 36.5 from 26.2 for V4-Flash-0731
- Terminal Bench 2.1 score of 83.9 versus 85.0 for Opus-4.8
- MIT-licensed weights include minimal PyTorch inference and DSpark attention details
- Served via vLLM on a single 4xGB300 node with FP8 KV cache
- Tags
- ai-agentsdeepseekdeepseek-v4-flash-vision-expmultimodalopen-weightssglangvision-language-modelvllm
- Vendors
- DeepSeek
- Products
- DeepSeek-V4-Flash-Vision-Exp
- Organizations
- DeepSeek
- AI models
- DeepSeek-V4-Flash-Vision-Exp
From the original article
3 min read at huggingface.coimage-text-to-text · 313,547 downloads · 812 likes · license mit
By deepseek-ai · huggingface.co
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.