rfdetr-onnx

RF-DETR exported to ONNX for use as ffrwd model pins. The network is Roboflow's RF-DETR, Apache-2.0 for the Nano through Large sizes. This repo holds faithful exports of four checkpoints, nothing more: two of Roboflow's own, one third-party fine-tune for faces, and one fine-tune of our own for licence plates.

All are fp32, exported by the rfdetr pip package's own model.export(format="onnx") at each variant's own training size.

file	variant	checkpoint	input	outputs
rf-detr-large-2026.onnx	RF-DETR Large (2026), 33.9M params, 56.5 COCO AP	rf-detr-large-2026.pth	1x3x704x704	dets 1x300x4, labels 1x300x91
rf-detr-seg-large.onnx	RF-DETR Seg Large, 36.2M params, 47.1 COCO AP	rf-detr-seg-large.pt	1x3x504x504	dets 1x200x4, labels 1x200x91, masks 1x200x126x126
rf-detr-medium-face.onnx	RF-DETR Medium fine-tuned for faces, one class	rfdetr_medium_face.pth	1x3x576x576	dets 1x300x4, labels 1x300x1
rf-detr-medium-plate.onnx	RF-DETR Medium fine-tuned for licence plates, one class	rfdetr_medium_plate.pth	1x3x576x576	dets 1x300x4, labels 1x300x2

Hashes

file	bytes	sha256
rf-detr-large-2026.onnx	128189032	da7f1bc118c6071b48096149b5df01eb1caeb3bb60a262949ac0e41f7c96f705
rf-detr-seg-large.onnx	137933860	e6154493b7ba38c285f416db812589267a7292a736fa8564577bcd27a9012e9e
rf-detr-medium-face.onnx	127032882	b594500a99c3a4287805b7afa15eb74ee559be937eb00315768612846f93cc45
rf-detr-medium-plate.onnx	126302012	2597bc5f0f71e2a6c8da3983367d774599d0e587b2e298238fd8fb71bd1af337

The checkpoints each export started from. The two official ones are as the rfdetr package fetched them into ~/.roboflow/models; the face one is as downloaded from its repo; the plate one is the EMA weights our training run wrote, kept by hash rather than here.

checkpoint	bytes	sha256
rf-detr-large-2026.pth	135954129	0f4e20e19a99c0f8a62b5685f57f6c8b5c371c59081feda6752a0561a79ccf38
rf-detr-seg-large.pt	145055866	ca7b7c630ba22496067cc4f034c4e70c8e47fd7adcea04c3a49aa8c1755cbe6b
rfdetr_medium_face.pth	133666287	7e2d5e5a4087ecf98a73685679ddb0d667dc6da512be29c0fba5469816ebfe06
rfdetr_medium_plate.pth	133776283	6d4c2821ede40269e83f694f474d137d0e67d066880ead09ec51abd0b864fcc4

The two COCO exports were made with rfdetr==1.10.0, the face and plate exports with rfdetr==1.10.1 (scripts/export_face.py, scripts/export_plate.py), each with torch from the package's own dependency set, opset 17, IR 8. The exporter does not simplify the graph: onnxsim is a listed dependency it never calls, so what is here is the graph as exported.

Model I/O

The input is planar RGB in NCHW, the whole frame stretched to the square with no aspect padding, scaled to 0..1 and then normalized with ImageNet's mean (0.485, 0.456, 0.406) and standard deviation (0.229, 0.224, 0.225). The graphs do not bake that in: their first node reads the input tensor directly.

dets is cx, cy, w, h, each a fraction of the picture. labels is raw class logits, so a confidence is sigmoid of the logit. For the two COCO exports the logits are in COCO's 91-slot category numbering - slot 1 is person, 62 chair, 72 tv - and a query's class is the largest of them. masks is raw mask logits per query over the whole square at quarter resolution; a pixel is inside its instance where the logit crosses zero. A DETR head returns no duplicates, so there is no NMS.

The face export

rf-detr-medium-face.onnx is RF-DETR Medium (DINOv2 backbone) as fine-tuned for single-class face detection by Herojayjay/RFDETR-Face-Detection, Apache-2.0, on a Kaggle face dataset of about 16,700 annotated images, at 576x576. It is not Roboflow's checkpoint and not trained on WIDER FACE; the WIDER FACE figures below are a check, not its benchmark.

Its head is one logit wide: labels is 1x300x1, index 0 is the face, and there is no background slot. The checkpoint's own class_embed is (1, 256), so the export loads it with num_classes left unset and lets the package align the head to the checkpoint. Passing num_classes=1 would tile the trained row into a second, invented class.

Measured on 200 WIDER FACE validation images (3,378 annotated faces) at a single 576 pass, matching at IoU 0.5, face size as the square root of the box area on the original image:

confidence	recall	precision	recall under 32 px	32-96 px	over 96 px
0.25	0.32	0.87	0.17	0.82	0.98
0.50	0.21	0.97	0.07	0.71	0.98

Small faces dominate that benchmark (2,649 of the 3,378 are under 32 px), and a single 576 pass cannot resolve them; the medium and large columns are the figures that describe a face that fills more of the frame.

The plate export

rf-detr-medium-plate.onnx is RF-DETR Medium fine-tuned by us for single-class licence plate detection, at 576x576, from Roboflow's rf-detr-medium.pth with rfdetr==1.10.1, on Open Images V7's "Vehicle registration plate" boxes (annotations CC BY 4.0, images CC BY 2.0) with the group-of boxes dropped: 4,497 training images carrying 6,554 plates, 724 validation images carrying 985. Twenty epochs, batch 8 with two steps of gradient accumulation, on one RTX 4090; validation mAP 50:95 peaked at 0.655 after the fourth epoch and the EMA weights of that epoch are what was exported (scripts/export_plate.py). The training data is what the model knows: photographs, mostly of parked and passing cars, plates from many countries, few of them at night.

Its head is two logits wide: rfdetr 1.10.1 lays a one-class head out as the class plus one spare slot, so labels is 1x300x2, index 0 is the plate, and index 1 was never trained toward anything and stays quiet - on the 2,065 test images below not one query's spare logit cleared 0.5. There is no background slot to subtract.

Measured on Open Images' own test split, 2,065 images holding 2,835 plates, at a single 576 pass, matching at IoU 0.5, plate size as the box's width on the original image:

confidence	recall	precision	recall under 32 px	32-96 px	over 96 px
0.25	0.87	0.91	0.47	0.96	0.98
0.50	0.80	0.98	0.23	0.90	0.97

Of the 2,835 plates, 580 are under 32 px wide, and a single 576 pass cannot resolve most of them; the two right-hand columns describe a plate that fills more of the frame.

Not here

RF-DETR XLarge and 2XLarge, and their segmentation counterparts, are under Roboflow's PML 1.0 rather than Apache-2.0, and are not exported here. The size names changed in 2026: the pre-rename "large" (487 MB fp32, 126M params) is today's XLarge, so check parameter counts and file sizes rather than names.