DeepSeek-V4-Flash-Vision-Exp

DeepSeek-V4-Flash-Vision-Exp is an experimental multimodal model that extends the DeepSeek-V4-Flash architecture with visual modules for enhanced visual understanding. It maintains high-performance text-only agent capabilities while significantly improving multimodal agent performance across various benchmarks.

DeepSeek-V4-Flash-Vision-Exp is an experimental multimodal model that extends the DeepSeek-V4-Flash architecture with visual modules for enhanced visual understanding. It maintains high-performance text-only agent capabilities while significantly improving multimodal agent performance across various benchmarks.

• 01Multimodal input support for text and images
• 02Maintains high-performance text-only agent capabilities
• 03Optimized for agentic workflows and tool-calling
• 04Open-weights release under MIT license
• 05Compatible with vLLM and SGLang for efficient inference
• 06Supports speculative decoding via DSpark