DeepSeek V4 Flash Vision Uncensored 3.87bpw Q3 K needs about 154.6 GB of memory at Q3_K (a 137.6 GB file) with an 8K context window, which is more than a 48 GB GPU holds.
DeepSeek V4 Flash Vision Uncensored 3.87bpw Q3_K – large‑context vision‑enabled LLM
DeepSeek V4 Flash Vision Uncensored 3.87bpw Q3_K is a 284‑billion‑parameter model quantized for llama.cpp. It supports a 1,048,576‑token context and retains the original vision tower. The quantization uses 3.87 bits per weight and is designed for high‑memory GPUs.
Good at
- Large context reasoning
- Vision‑enabled inference
- High‑throughput decoding
Limits
- Requires ~154.6 GB RAM at 8 k context
- No license information provided
- Only Q3_K quant available
Best for
- Research on long‑form vision tasks
- Enterprise inference on high‑memory GPUs
- Benchmarking large‑context LLMs
Who it’s for. Researchers and developers needing a vision‑capable LLM with a very long context window. Suitable for teams with access to 128 GB RAM and 16–24 GB VRAM GPUs.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q3_K | 137.6 GB | 154.6 GB | Too big | Get file → |
The Q3_K quant is the only variant available. It requires at least 154.6 GB RAM for 8 k context, so it is best suited for machines with 128 GB RAM or more. If your system has less memory, this model is not recommended.
What GPU do you need for DeepSeek V4 Flash Vision Uncensored 3.87bpw Q3 K?
These figures add the file size, an fp16 KV cache at an 8K context window and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: no quant fits at an 8K context.
- 16 GB VRAM: no quant fits at an 8K context.
- 24 GB VRAM: no quant fits at an 8K context.
- 32 GB VRAM: no quant fits at an 8K context.
- 48 GB VRAM: no quant fits at an 8K context.
Which DeepSeek V4 Flash Vision Uncensored 3.87bpw Q3 K quant should you choose?
Q3_K (137.6 GB) is the middle option in this repository. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Frequently asked questions about DeepSeek V4 Flash Vision Uncensored 3.87bpw Q3 K
How much VRAM does DeepSeek V4 Flash Vision Uncensored 3.87bpw Q3 K need?
About 154.6 GB at Q3_K with an 8K context window. That is the 137.6 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run DeepSeek V4 Flash Vision Uncensored 3.87bpw Q3 K on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 154.6 GB, so you would need a larger GPU or CPU offloading.
What is the file size of DeepSeek V4 Flash Vision Uncensored 3.87bpw Q3 K GGUF?
It depends on the quant: Q3_K is 137.6 GB.
What context length does DeepSeek V4 Flash Vision Uncensored 3.87bpw Q3 K support?
The GGUF header lists a trained context length of 1,048,576 tokens (1024K).
Where can I download DeepSeek V4 Flash Vision Uncensored 3.87bpw Q3 K GGUF?
The files are on Hugging Face in the MetaphoricalCode/DeepSeek-V4-Flash-Vision-Uncensored-3.87bpw-Q3_K-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
Can I run this model on 8 GB VRAM?
No. The quant requires at least 16 GB VRAM and 128 GB RAM to run efficiently.
Does this model support vision tasks?
Yes, the vision tower remains intact, so it can process images.
What is the context length of this model?
It supports up to 1,048,576 tokens.