vllm-d3-mini needs about 5.3 GB of memory at Q4_K_M (a 3.7 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 3 quants, from 3.7 GB (Q4_K_M) to 9.7 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.
4B multimodal decision model for text, image, and video queries
vllm-d3-mini is a 4.8 billion‑parameter multimodal foundation model. It answers multiple choice, yes/no, and scoring questions in a single call, returning probabilities for each option. The model supports text, images, and videos, and is optimized for decision‑making tasks.
Good at
- Fast single‑call decision answers
- Multimodal input handling
- Probability‑based outputs
Limits
- Requires 8 k context for memory estimates
- No text generation, only decision outputs
- Licensed under Apache‑2.0
Best for
- Rapid decision‑making in applications
- Multimodal QA systems
- Batching many questions per request
Who it’s for. Developers building decision‑oriented AI services that need quick, probability‑based answers from text, images, or videos. The model is suitable for internal tools and APIs that require deterministic choice outputs.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 3.7 GB | 5.3 GB | Fits | Get file → | |
| Q8_0 | 5.2 GB | 6.7 GB | Fits | Get file → | |
| BF16 | 9.7 GB | 11.3 GB | Too big | Get file → |
For 8 GB VRAM, use Q4_K_M (3.71 GB, ~5.3 GB RAM). For 12 GB VRAM, Q8_0 (5.16 GB, ~6.7 GB RAM). BF16 requires 9.7 GB and ~11.3 GB RAM, suitable for larger GPUs.
What GPU do you need for vllm-d3-mini?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) 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: the best quant that fits is Q4_K_M (3.7 GB file, about 5.3 GB in use).
- 8 GB VRAM: the best quant that fits is Q8_0 (5.2 GB file, about 6.7 GB in use).
- 12 GB VRAM: the best quant that fits is BF16 (9.7 GB file, about 11.3 GB in use).
Which vllm-d3-mini quant should you choose?
Q4_K_M is the usual starting point for vllm-d3-mini: it balances file size and quality well. Here it is a 3.7 GB file.
The Q8_0 file is 5.2 GB, about 1.4 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about vllm-d3-mini
How much VRAM does vllm-d3-mini need?
About 5.3 GB at Q4_K_M with an 8K context window. That is the 3.7 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 vllm-d3-mini on an 8 GB GPU?
Yes, with Q8_0: the 5.2 GB file needs about 6.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of vllm-d3-mini GGUF?
It depends on the quant: Q4_K_M is 3.7 GB, Q8_0 is 5.2 GB, BF16 is 9.7 GB.
What context length does vllm-d3-mini support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.
Where can I download vllm-d3-mini GGUF?
The files are on Hugging Face in the ggml-org/vllm-d3-mini-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does vllm-d3-mini use?
Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run vllm-d3-mini on 8 GB VRAM?
Yes, the Q4_K_M quantized file is 3.71 GB and requires about 5.3 GB of RAM at 8 k context.
Does this model support text generation?
No, it only returns decision probabilities for choice, yes/no, or score questions.
What license applies to this model?
The model is released under the Apache‑2.0 license.