Add/update model card for nano

README.md

@@ -0,0 +1,74 @@
+---
+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 Nano — GGUF for rfdetr.cpp
+
+GGUF-format weights of [Roboflow RF-DETR Nano](https://github.com/roboflow/rf-detr) (detection variant) for use with [rfdetr.cpp](https://github.com/mudler/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-nano-f32.gguf` | F32 | 112.7 | 0.9890 | 0.9890 | 0.0068 | 55.8 |
+| `rfdetr-nano-f16.gguf` ← **recommended** | F16 | 60.5 | 0.9890 | 0.9890 | 0.0070 | 49.6 |
+| `rfdetr-nano-q8_0.gguf` | Q8_0 | 36.0 | 0.9890 | 0.9890 | 0.0084 | 59.3 |
+| `rfdetr-nano-q4_K.gguf` | Q4_K | 29.7 | 0.9162 | 0.7585 | 0.0303 | 60.6 |
+
+All accuracy numbers are computed against the upstream PyTorch reference (`rfdetr 1.7.0`) on 7 COCO val2017 images at threshold 0.5. Latency is measured with `rfdetr-cli bench` (8 iters + 3 warmup) at T=8 threads on a single AMD Ryzen 9 9950X3D image (`coco_kitchen.jpg`, 640x427).
+
+## Architecture
+
+- Backbone: DINOv2-small
+- Input resolution: 384×384
+- Patch size: 14
+- Decoder layers: 2
+- 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).
+
+## Usage
+
+```bash
+# 1. Clone + build rfdetr.cpp
+git clone https://github.com/mudler/rf-detr.cpp
+cd rt-detr.cpp
+cmake -B build -DRFDETR_BUILD_CLI=ON && cmake --build build -j
+
+# 2. Download a quant (F16 recommended)
+hf download mudler/rfdetr-cpp-nano rfdetr-nano-f16.gguf --local-dir models/
+
+# 3. Run detection
+build/bin/rfdetr-cli detect \
+    --model models/rfdetr-nano-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 1.7.0) on 7 COCO val2017 images 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/mudler/rf-detr.cpp/blob/main/BENCHMARK.md) and [`benchmarks/results/accuracy_sweep.json`](https://github.com/mudler/rf-detr.cpp/blob/main/benchmarks/results/accuracy_sweep.json) for the full sweep across all 32 (variant × quant) cells.
+
+## License
+
+Apache-2.0 — matches the upstream [rfdetr](https://github.com/roboflow/rf-detr) license.