vllm-d3-nano needs about 3.5 GB of memory at Q8_0 (a 2.6 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 2 quants, from 2.6 GB (Q8_0) to 4.8 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 1.6 GB.
vllm-d3-nano: 2.4B multimodal decision model
vllm-d3-nano is a 2.4B parameter Qwen‑35 architecture model that can process text, images, and videos. It answers multiple-choice, yes/no, and scoring questions in a single call, returning probabilities for each option. The model is licensed under Apache‑2.0.
Good at
- Multimodal decision tasks
- Fast inference on single GPU
- Single‑call multi‑question responses
Limits
- No text generation
- Requires 8k context for memory estimates
- No explicit language support listed
Best for
- Decision‑making APIs
- Rapid multimodal QA
- Batching multiple-choice queries
Who it’s for. Developers building decision‑oriented services that need to evaluate text, images, and videos together. The model is suitable for applications requiring probability‑based answers without text generation.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q8_0 | 2.6 GB | 3.5 GB | Fits | Get file → | |
| BF16 | 4.8 GB | 5.7 GB | Fits | Get file → |
For 8 GB VRAM, use the Q8_0 quantization (file size 2.55 GB, estimated 3.5 GB memory at 8k context). If more VRAM is available, BF16 (4.79 GB, 5.7 GB memory) offers higher precision.
What GPU do you need for vllm-d3-nano?
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: the best quant that fits is Q8_0 (2.6 GB file, about 3.5 GB in use).
- 6 GB VRAM: the best quant that fits is BF16 (4.8 GB file, about 5.7 GB in use).
Which vllm-d3-nano quant should you choose?
Q8_0 (2.6 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about vllm-d3-nano
How much VRAM does vllm-d3-nano need?
About 3.5 GB at Q8_0 with an 8K context window. That is the 2.6 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 vllm-d3-nano on an 8 GB GPU?
Yes, with BF16: the 4.8 GB file needs about 5.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of vllm-d3-nano GGUF?
It depends on the quant: Q8_0 is 2.6 GB, BF16 is 4.8 GB.
What context length does vllm-d3-nano support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.4 GB at 8K and 1.6 GB at 32K.
Where can I download vllm-d3-nano GGUF?
The files are on Hugging Face in the ggml-org/vllm-d3-nano-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-nano 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-nano on 8 GB VRAM?
Yes, the Q8_0 quantized model requires about 3.5 GB of VRAM at 8k context, which fits within an 8 GB GPU.
Does vllm-d3-nano support text generation?
No, the model is designed for decision tasks and does not generate text.
What license applies to vllm-d3-nano?
The model is released under the Apache‑2.0 license.