CoreEpoch/rtdetrv2-s-int8-onnx

RT-DETRv2-S: INT8 ONNX, 45.7 AP COCO in 32.7 MB

RT-DETRv2-S (PekingU/rtdetr_v2_r18vd, Apache-2.0) quantized with Kenosis: 128-image COCO calibration, no retraining. 45.7 AP (95.0% of FP32) from a single 32.7 MB file, on ONNX Runtime or OpenVINO, CPU or GPU, no accelerator required.

Accuracy

COCO val2017, 4,800 images (disjoint from the 128 calibration images), pycocotools bbox.

For licensing inquiries or custom model quantization: coreepoch.dev/kenosis · licensing@coreepoch.dev

Run

from huggingface_hub import hf_hub_download
import numpy as np, onnxruntime as ort
from PIL import Image
config = hf_hub_download("CoreEpoch/rtdetrv2-s-int8-onnx", "config.json")
path = hf_hub_download("CoreEpoch/rtdetrv2-s-int8-onnx", "rtdetrv2_s_640_int8_kenosis.onnx")
sess = ort.InferenceSession(path, providers=["CPUExecutionProvider"])
img = Image.open("your_image.jpg").convert("RGB")
x = np.asarray(img.resize((640, 640), Image.BILINEAR), np.float32) / 255.0
logits, boxes = sess.run(None, {"input": x.transpose(2,0,1)[None]})
scores = 1.0 / (1.0 + np.exp(-logits[0]))
q, c = divmod(int(scores.argmax()), scores.shape[1])
print(f"class {c}, score {scores.max():.2f}, box {boxes[0][q]}")

Input 1x3x640x640, RGB, /255 (no mean/std). Outputs logits [1,300,80] (contiguous COCO-80 order) and boxes [1,300,4] (normalized cxcywh).

Integrity & license

rtdetrv2_s_640_int8_kenosis.onnx (32,666,936 B), SHA-256 2FBEA12F3A66DA22BE0C0404F78D378F0D6F4C2F8107A5444EE00C6AB717974D. Apache-2.0 (base RT-DETRv2, Lv et al. 2024, retained).

Notice: The open-source license applies strictly to the published model weights and ONNX inference graph. Core Epoch LLC explicitly reserves all proprietary rights, including patent, copyright, and trade secret rights, in the Kenosis quantization software, graph compilation toolchain, and underlying algorithmic optimization processes. No compiler license is granted or implied.