Mirrored from deepseek-ai/DeepSeek-V4-Flash-Vision-Exp
All models by deepseek-ai →
We are excited to introduce DeepSeek-V4-Flash-Vision-Exp, our first experimental multimodal model in the DeepSeek-V4 family. It builds on the DeepSeek-V4-Flash architecture by incorporating visual modules and undergoing continued training to unlock visual understanding capabilities.
Compared to DeepSeek-V4-Flash-0731, DeepSeek-V4-Flash-Vision-Exp achieves substantial improvements on its multimodal agent capabilities, while maintaining comparable performance on text-only agent tasks.
Notes:
max reasoning effort level with temperature = 1.0, top_p = 0.95.This repository contains the tokenizer, prompt encoding reference, and a minimal PyTorch inference implementation for DeepSeek-V4 Flash Vision. The reference inference covers the vision encoder and aligner, DFlash attention, MoE, Hyper-Connections, and the DSpark forward path.
encoding/ and inference/ deliberately remain separate: prompt formatting
does not depend on PyTorch, while inference imports the sibling encoding module
with an explicit Python path. No symlinks are required.
The tokenizer files are regular files so that the repository can be uploaded
to Hugging Face without relying on local filesystem symlinks. The large model
shards are described by model.safetensors.index.json and are not duplicated
inside the source checkout used to assemble this repository.
See encoding/README.md [blocked]. Both OpenAI-style JSON content
blocks and the compact <image>path</image> TXT notation are supported. The two
examples under inference/examples/ encode to identical prompts and token IDs.
See inference/README.md [blocked] for dependency installation,
checkpoint conversion, and TXT/JSON inference commands.
For example, the command below serves the model with vLLM on a single 4×GB300 node. See the vLLM recipe for detailed instructions and other hardware configurations.
Enable DSpark with --speculative-algorithm DSPARK and do not set a separate --speculative-draft-model-path as the target and draft weights therefore come from the same checkpoint. See the SGLang cookbook for detailed instructions, benchmarks and other hardwares configurations.
This repository is licensed under the MIT License [blocked].
.├── encoding/ # OpenAI-style messages -> model prompt├── inference/ # weight conversion and minimal inference│ └── examples/ # equivalent TXT and JSON vision prompts├── config.json # Hugging Face model metadata├── generation_config.json├── model.safetensors.index.json├── tokenizer.json└── tokenizer_config.jsondocker run --gpus all \ vllm/vllm-openai:deepseekv4-flash-vision deepseek-ai/DeepSeek-V4-Flash-Vision-Exp \ --kv-cache-dtype fp8 \ --block-size 256 \ --tensor-parallel-size 4 \ --tool-call-parser deepseek_v4 \ --enable-auto-tool-choice \ --reasoning-parser deepseek_v4 \ --reasoning-config '{"reasoning_parser":"deepseek_v4","reasoning_start_str":"","reasoning_end_str":""}' \ --speculative-config '{"method":"dspark","model":"deepseek-ai/DeepSeek-V4-Flash-Vision-Exp","num_speculative_tokens":3,"draft_sample_method":"probabilistic","enable_adaptive_verification":true}'sglang serve \ --model-path deepseek-ai/DeepSeek-V4-Flash-Vision-Exp \ --tp 4 \ --speculative-algorithm DSPARK \ --mem-fraction-static 0.85 \ --host 0.0.0.0 \ --port 30000