rfdetr cpp nano needs about 0.5 GB of memory at Q8_0 (a 0.0 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 4 quants, from 0.0 GB (Q4_K) to 0.1 GB (F32).
RF-DETR Nano – lightweight object detection model in GGUF format
RF-DETR Nano is a compact detection variant based on DINOv2-small. It runs on CPU with rfdetr.cpp and supports four quantizations. The model delivers box-only detections with a 384×384 input resolution.
Good at
- Fast CPU inference
- Small model size
- Accurate COCO detection
Limits
- No vision support beyond boxes
- Limited to 8k context
- Requires CPU, not GPU optimized
Best for
- Embedded inference
- Edge devices
- CPU‑only deployments
Who it’s for. Developers needing a lightweight detection model for CPU‑only environments. The model is suitable for edge devices, mobile inference, and quick prototyping where GPU resources are unavailable.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K | 0.0 GB | 0.5 GB | Fits | Get file → | |
| Q8_0 | 0.0 GB | 0.5 GB | Fits | Get file → | |
| F16 | 0.1 GB | 0.6 GB | Fits | Get file → | |
| F32 | 0.1 GB | 0.6 GB | Fits | Get file → |
For 8 GB VRAM, the Q4_K file (0.03 GB) is ideal, using ~0.5 GB memory at 8k context. If higher accuracy is needed, choose Q8_0 (0.04 GB) or F16 (0.06 GB). F32 (0.12 GB) offers full precision but requires more memory.
What GPU do you need for rfdetr cpp nano?
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: the best quant that fits is F32 (0.1 GB file, about 0.6 GB in use).
Which rfdetr cpp nano quant should you choose?
Q8_0 (0.0 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.
If memory is tight, the smallest file is Q4_K at 0.0 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about rfdetr cpp nano
How much VRAM does rfdetr cpp nano need?
About 0.5 GB at Q8_0 with an 8K context window. That is the 0.0 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 rfdetr cpp nano on an 8 GB GPU?
Yes, with F32: the 0.1 GB file needs about 0.6 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of rfdetr cpp nano GGUF?
It depends on the quant: Q4_K is 0.0 GB, Q8_0 is 0.0 GB, F16 is 0.1 GB, F32 is 0.1 GB.
Where can I download rfdetr cpp nano GGUF?
The files are on Hugging Face in the mudler/rfdetr-cpp-nano repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does rfdetr cpp 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 RF‑DETR Nano on 8 GB VRAM?
Yes, the Q4_K quantization uses about 0.5 GB memory at 8k context, well within an 8 GB VRAM budget.
Which quantization gives the best speed on CPU?
F16 is recommended for fastest CPU performance while maintaining accuracy equal to F32.
Does the model support GPU inference?
The provided GGUF files are intended for CPU use with rfdetr.cpp; GPU inference is not supported.