Model Intelligence Sheet
ddh0/DeepSeek-V4-Flash-GGUF overview
GGUF quantizations of DeepSeek V4 Flash. Using MTP requires am17an/llama.cpp:dsv4 mtp https://github.com/am17an/llama.cpp/tree/dsv4 mtp until llama.cpp 25784 h…
Runs locally from ~3.51 GB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
5 GGUF files detected
Direct downloads for local inference
| File | Type | Quantization | Size | Link |
|---|---|---|---|---|
| DeepSeek-V4-Flash-3.86bpw-IQ3_S.gguf | GGUF | IQ3_S | 127.82 GB | Download |
| DeepSeek-V4-Flash-3.86bpw-Q3_K.gguf | GGUF | Q3_K | 127.82 GB | Download |
| DeepSeek-V4-Flash-MTP-3.93bpw.gguf | GGUF | GGUF | 3.51 GB | Download |
| DeepSeek-V4-Flash-MTP-4.93bpw.gguf | GGUF | GGUF | 4.41 GB | Download |
| DeepSeek-V4-Flash-MXFP4.gguf | GGUF | GGUF | 145.29 GB | Download |
Model Details
Model README
---
base_model:
- deepseek-ai/DeepSeek-V4-Flash
---
GGUF quantizations of DeepSeek-V4-Flash.
Using MTP requires am17an/llama.cpp:dsv4-mtp until llama.cpp#25784 is merged.
Run ddh0/DeepSeek-V4-Flash-GGUF with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models