Model Intelligence Sheet
ggml-org/DeepSeek-V4-Flash-Vision-Exp-GGUF overview
DeepSeek V4 Flash Vision Exp Run with https://llama.app bash llama serve hf ggml org/DeepSeek V4 Flash Vision Exp GGUF Source models https://huggingface.co/dee…
Runs locally from ~5.0 MB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
10 GGUF files detected
Direct downloads for local inference
| File | Type | Quantization | Size | Link |
|---|---|---|---|---|
| DeepSeek-V4-Flash-Vision-Exp-MXFP4-00001-of-00002.gguf | GGUF | GGUF | 5.0 MB | Download |
| DeepSeek-V4-Flash-Vision-Exp-MXFP4-00002-of-00002.gguf | GGUF | GGUF | 144.34 GB | Download |
| DeepSeek-V4-Flash-Vision-Exp-Q2_K-00001-of-00002.gguf | GGUF | Q2_K | 5.0 MB | Download |
| DeepSeek-V4-Flash-Vision-Exp-Q2_K-00002-of-00002.gguf | GGUF | Q2_K | 109.29 GB | Download |
| DeepSeek-V4-Flash-Vision-Exp-Q2_K_S-00001-of-00002.gguf | GGUF | Q2_K_S | 5.0 MB | Download |
| DeepSeek-V4-Flash-Vision-Exp-Q2_K_S-00002-of-00002.gguf | GGUF | Q2_K_S | 91.82 GB | Download |
| dspark-DeepSeek-V4-Flash-Vision-Exp-BF16.gguf | GGUF | BF16 | 10.15 GB | Download |
| dspark-DeepSeek-V4-Flash-Vision-Exp-MXFP4.gguf | GGUF | GGUF | 10.08 GB | Download |
| mmproj-DeepSeek-V4-Flash-Vision-Exp-BF16.gguf | GGUF | BF16 | 891.2 MB | Download |
| mmproj-DeepSeek-V4-Flash-Vision-Exp-Q8_0.gguf | GGUF | Q8_0 | 474.9 MB | Download |
Model Details
| Model ID | ggml-org/DeepSeek-V4-Flash-Vision-Exp-GGUF |
|---|---|
| Author | ggml-org |
| Pipeline | image-text-to-text |
| License | mit |
| Base model | deepseek-ai/DeepSeek-V4-Flash-Vision-Exp |
| Last modified | 2026-09-02T21:00:42.000Z |
Model README
---
license: mit
pipeline_tag: image-text-to-text
tags:
- gguf
- quantized
base_model:
- deepseek-ai/DeepSeek-V4-Flash-Vision-Exp
---
DeepSeek-V4-Flash-Vision-Exp
Run with https://llama.app
llama serve -hf ggml-org/DeepSeek-V4-Flash-Vision-Exp-GGUFSource models
- https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-Vision-Exp
Notes
- Includes a Q8_0 mmproj for the vision encoder.
- Currently, the Q2 models do not use an imatrix calibration due to lack of one.
TODOs
- add info
> [!IMPORTANT]
> This model is automatically converted using https://github.com/ggml-org/convert
Run ggml-org/DeepSeek-V4-Flash-Vision-Exp-GGUF with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models