File size: 5,359 Bytes
7307975 1eff059 7307975 b2d88f9 7307975 1eff059 7307975 b2d88f9 7307975 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 | ---
license: apache-2.0
library_name: rfdetr.cpp
tags:
- object-detection
- rfdetr
- gguf
- ggml
- cpp-inference
pipeline_tag: object-detection
base_model: roboflow/rfdetr
---
# RF-DETR Medium — GGUF for rfdetr.cpp
GGUF-format weights of [Roboflow RF-DETR Medium](https://github.com/roboflow/rf-detr) (detection variant) for use with [rfdetr.cpp](https://github.com/adithyab94/rf-detr.cpp), a C++/ggml implementation that matches the upstream PyTorch model on CPU.
This repo contains all four standard quantizations of this variant. **F16 is the recommended default** — same accuracy as F32, 1.85× smaller, and typically the fastest on modern CPUs thanks to ggml's F32×F16 matmul fast path.
## Available files
| File | Quant | Size (MB) | Recall @ IoU 0.5 | Recall @ IoU 0.95 | Mean \|Δscore\| | Latency (median ms, T=8) |
|---|---|---:|---:|---:|---:|---:|
| `rfdetr-medium-f32.gguf` | F32 | 125.0 | 0.9731 | 0.9621 | 0.0024 | 139.9 |
| `rfdetr-medium-f16.gguf` ← **recommended** | F16 | 67.2 | 0.9780 | 0.9670 | 0.0022 | 143.9 |
| `rfdetr-medium-q8_0.gguf` | Q8_0 | 40.2 | 0.9890 | 0.9691 | 0.0070 | 145.8 |
| `rfdetr-medium-q4_K.gguf` | Q4_K | 32.5 | 0.8828 | 0.7390 | 0.0199 | 155.8 |
All accuracy numbers above are computed against the upstream PyTorch reference (`rfdetr 1.9.0`) on 7 images (000000000139.jpg, 000000000632.jpg, 000000039769.jpg, 000000087038.jpg, 000000252219.jpg, 000000397133.jpg, bus.jpg) at threshold 0.5. Latency is measured separately with `rfdetr-cli bench` (8 iters + 3 warmup) at T=8 threads on a single Intel Core i7-12800HX, on `tests/fixtures/ci/test_image.jpg`.
## Architecture
- Backbone: DINOv2-small
- Input resolution: 576×576
- Patch size: 16
- Decoder layers: 4
- Object queries: 300
- Task: object detection (boxes only)
## Quantization notes
- **F32** — full-precision reference, ~120 MB. Bit-exact PyTorch parity.
- **F16** — matmul-multiplicand weights only; LayerNorms, conv kernels, embeddings, biases, and layer-scale gammas stay F32. Lossless on this model and consistently the fastest variant on CPU.
- **Q8_0** — best size/accuracy tradeoff under F16; ~3× smaller than F32 with effectively identical detections.
- **Q4_K** — smallest practical quant. Rows with `ne[0] % 256 != 0` (the decoder's 128-dim MLP halves, 60 tensors) silently fall back to Q8_0 per ggml's quantizer logic — net compression is still ~3.8× over F32. Use only when the size budget is tight; expect a measurable Recall@0.95 drop relative to F16/Q8_0 (see file table above).
## Compatibility
These GGUFs stamp `rfdetr.preprocess.resize_mode = "bilinear_no_antialias"`, matching RF-DETR 1.9's antialias-free float bilinear resize (`align_corners=false`, half-pixel coordinates, no intermediate uint8 rounding). rf-detr.cpp treats this key as **optional**: GGUFs that predate this metadata (no `resize_mode` key) keep using the legacy stb-based resize path, so older files continue to produce their original outputs unchanged. An unrecognized `resize_mode` value is rejected rather than guessed.
**Keypoint-preview inference is not supported.** rf-detr.cpp does not implement the keypoint output head; this repository only serves box detection outputs.
## Usage
```bash
# 1. Clone + build rfdetr.cpp
git clone https://github.com/adithyab94/rf-detr.cpp
cd rf-detr.cpp
cmake -B build -DRFDETR_BUILD_CLI=ON && cmake --build build -j
# 2. Download a quant (F16 recommended)
hf download adithya-balaji/rfdetr-cpp-medium rfdetr-medium-f16.gguf --local-dir models/
# 3. Run detection
build/bin/rfdetr-cli detect \
--model models/rfdetr-medium-f16.gguf \
--input my_image.jpg \
--threshold 0.5 --threads 8 \
--output detections.json
```
## Accuracy methodology
All accuracy metrics are computed against the upstream PyTorch reference (`rfdetr 1.9.0`) on 7 images (000000000139.jpg, 000000000632.jpg, 000000039769.jpg, 000000087038.jpg, 000000252219.jpg, 000000397133.jpg, bus.jpg) at threshold 0.5. Each detection match uses greedy Hungarian-style assignment by IoU (≥ 0.5 lenient, ≥ 0.95 strict) with class equality required.
See [BENCHMARK.md](https://github.com/adithyab94/rf-detr.cpp/blob/main/BENCHMARK.md) and [`benchmarks/results/accuracy_sweep.json`](https://github.com/adithyab94/rf-detr.cpp/blob/main/benchmarks/results/accuracy_sweep.json) for the full sweep across the (variant × quant) cells.
## Provenance
- Source project: [Roboflow RF-DETR](https://github.com/roboflow/rf-detr)
- Upstream package: `rfdetr==1.9.0`
- Converted with [rfdetr.cpp](https://github.com/adithyab94/rf-detr.cpp) at commit [`fbef9387bed3`](https://github.com/adithyab94/rf-detr.cpp/commit/fbef9387bed3f3c82bf3f4e041253ffe3081f275)
- Checkpoint: official pretrained `rfdetr-medium` weights (downloaded by the `rfdetr` package on first use)
### Checksums (SHA-256)
Also available as `SHA256SUMS` in this repo.
```
664ffe389100cbc942a5d99a2d5b48ef12fdf88f047471fd42d98c5c7d8d32fb rfdetr-medium-f32.gguf
353bdccb3a6aa4865fe5c6b636d627c7d6f8c6f94701b9d074d5c2b757b4c792 rfdetr-medium-f16.gguf
b8f77932ceb8429856588aabfa415a5a752eba2d6c212b043faf473a8aab575f rfdetr-medium-q8_0.gguf
7a6c0da4b1b1070986f30582eb8b1b7439b386b15a2171707346eb7c402cca60 rfdetr-medium-q4_K.gguf
```
## License
Apache-2.0 — matches the upstream [rfdetr](https://github.com/roboflow/rf-detr) license.
|