Model pricing

Credit prices for the models served by the Serverless Cloud API, per 1,000 images.

Serverless Cloud API image inference is billed in credits per image. Every price on this page is credits per 1,000 images at the standard rate, rounded to four decimal places. Each row is a model architecture priced from its smallest size, and each section expands to show every size.

Volume discounts apply automatically once monthly usage crosses a threshold. Pricing covers how billing works, including Workflow runs and which failed requests are billable. Video is priced separately in Serverless Video Streaming.

Object detection

Sorted by credits per 1,000 images divided by COCO mAP 50:95, using the smallest size of each architecture. Accuracy figures come from the Roboflow model leaderboard.

These are the five most efficient architectures for this task.

Model	Credits per 1,000 images	Sizes	Best COCO mAP 50:95
RF-DETR	0.125	7	59.9%
YOLO26	0.125	5	56.3%
YOLOv12	0.125	5	54.0%
YOLOv11	0.125	5	53.6%
YOLOv10	0.125	6	53.6%

Every object detection model and size

• RF-DETR

  • Nano: 0.125

  • Small, Base, Medium, Large: 0.1875

  • XL: 0.25

  • 2XL: 0.3125

• YOLO26

  • Nano: 0.125

  • Small, Medium, Large: 0.1875

  • XL: 0.25

• YOLOv12

  • Nano: 0.125

  • Small, Medium, Large: 0.1875

  • XL: 0.25

• YOLOv11

  • Nano: 0.125

  • Small, Medium, Large: 0.1875

  • XL: 0.25

• YOLOv10

  • Nano: 0.125

  • Small, Medium, Balanced, Large: 0.1875

  • XL: 0.25

• YOLOv8

  • Nano: 0.125

  • Small, Medium, Large: 0.1875

  • XL: 0.25

• YOLOv9

  • Tiny: 0.125

  • Small, Medium, Compact: 0.1875

  • Extended: 0.25

• YOLOv7

  • Tiny: 0.125

  • Base: 0.1875

  • X: 0.25

• YOLOv5

  • Nano: 0.125

  • Small, Medium, Large: 0.1875

  • XL: 0.25

• YOLOLite

  • Nano: 0.125

  • Small, Medium, Large: 0.1875

  • XL: 0.25

• YOLO-NAS

  • Small, Medium: 0.1875

Instance segmentation

Sorted by price, lowest first.

These are the five lowest-priced architectures for this task.

Model	Credits per 1,000 images	Sizes
RF-DETR Seg	0.1875	7
YOLO26 Seg	0.1875	5
YOLOv11 Seg	0.1875	5
YOLOv8 Seg	0.1875	5
YOLOv5 Seg	0.1875	5

Every instance segmentation model and size

• RF-DETR Seg

  • Nano: 0.1875

  • Small, Medium, Large, Preview: 0.25

  • XL, 2XL: 0.3125

• YOLO26 Seg

  • Nano: 0.1875

  • Small, Medium, Large: 0.25

  • XL: 0.3125

• YOLOv11 Seg

  • Nano: 0.1875

  • Small, Medium, Large: 0.25

  • XL: 0.3125

• YOLOv8 Seg

  • Nano: 0.1875

  • Small, Medium, Large: 0.25

  • XL: 0.3125

• YOLOv5 Seg

  • Nano: 0.1875

  • Small, Medium, Large: 0.25

  • XL: 0.3125

• YOLACT

  • ResNet-101: 0.25

Semantic segmentation

Model	Credits per 1,000 images	Sizes
YOLO26 Semantic Seg	0.25	5
DeepLabV3+	0.25	1

Every semantic segmentation model and size

• YOLO26 Semantic Seg

  • Nano, Small, Medium, Large, XL: 0.25

• DeepLabV3+

  • MobileNetV2: 0.25

Keypoint detection

Model	Credits per 1,000 images	Sizes
YOLO26 Pose	0.125	5
YOLOv11 Pose	0.125	5
YOLOv8 Pose	0.125	5
RF-DETR Keypoint	0.25	1

Every keypoint detection model and size

• YOLO26 Pose

  • Nano: 0.125

  • Small, Medium, Large: 0.1875

  • XL: 0.25

• YOLOv11 Pose

  • Nano: 0.125

  • Small, Medium, Large: 0.1875

  • XL: 0.25

• YOLOv8 Pose

  • Nano: 0.125

  • Small, Medium, Large: 0.1875

  • XL: 0.25

• RF-DETR Keypoint

  • Preview: 0.25

Classification

Model	Credits per 1,000 images	Sizes
ViT	0.0625	1
ResNet	0.0625	4
DINOv3 Probe	0.0625	2

Every classification model and size

• ViT

  • B/16: 0.0625

• ResNet

  • 18, 34, 50, 101: 0.0625

• DINOv3 Probe

  • Small: 0.0625

  • Base: 0.125

Open-vocabulary detection

These models detect objects from a text prompt, with no training required.

Model	Credits per 1,000 images	Sizes
YOLO-World	0.1875	8
Roboflow Instant	0.75	1
Grounding DINO	0.75	1
OWLv2	0.75	2

Every open-vocabulary detection model and size

• YOLO-World

  • S, M, L, X, v2-S, v2-M, v2-L, v2-X: 0.1875

• Roboflow Instant

  • Single size: 0.75

• Grounding DINO

  • Swin-T: 0.75

• OWLv2

  • Large, Large (fine-tuned): 0.75

Segment Anything

Promptable segmentation foundation models.

Model	Credits per 1,000 images	Sizes
SAM 2	0.3125	4
SAM 3	0.5	2
SAM	0.5	1

Every model and size in Segment Anything

• SAM 2

  • Hiera Tiny, Hiera Small: 0.3125

  • Hiera Base+, Hiera Large: 0.5

• SAM 3

  • Base, Large: 0.5

• SAM

  • ViT-H: 0.5

OCR

PP-OCR is billed per run, not per stage. Calls with text detection turned off bill at the recognition-only rate, and every combination with detection on bills at the OCR rate.

Model	Credits per 1,000 images	Sizes
PP-OCR (recognition only)	0.125	3
TrOCR	0.3125	1
PP-OCR (detection + recognition)	0.375	3
DocTR	0.375	1
EasyOCR	0.375	1

Every model and size in OCR

• PP-OCR (recognition only)

  • Tiny, Small, Medium: 0.125

• TrOCR

  • Single size: 0.3125

• PP-OCR (detection + recognition)

  • Tiny, Small, Medium: 0.375

• DocTR

  • Single size: 0.375

• EasyOCR

  • Single size: 0.375

Embeddings

Model	Credits per 1,000 images	Sizes
CLIP	0.125	3
Perception Encoder	0.3125	1

Every embeddings model and size

• CLIP

  • ViT-B/16, ViT-B/32: 0.125

  • ViT-L/14: 0.1875

• Perception Encoder

  • Single size: 0.3125

Depth estimation

Model	Credits per 1,000 images	Sizes
Depth Anything V3	0.3125	2
Depth Anything V2	0.3125	1

Every depth estimation model and size

• Depth Anything V3

  • Small: 0.3125

  • Base: 0.75

• Depth Anything V2

  • Small: 0.3125

Execution time pricing

Custom Python blocks and vision language models are billed for the time they run, not per image. One credit buys 500 execution seconds.

Workload	Execution seconds per credit	Credits per second	Credits per hour
Custom Python blocks	500	0.002	7.2
Vision language models	500	0.002	7.2

Vision language models

Every vision language model is billed for its execution time at the rate above, whatever the prompt or the number of output tokens.

Model	Execution seconds per credit
PaliGemma 2	500
Florence 2	500
Qwen2.5 VL	500
Qwen3 VL	500
Qwen3.5	500
SmolVLM	500
Cosmos 3 (Edge) VLM	500
Moondream	500
Gemma 4	500
Mage-VL	500
GLM-OCR	500

A step that runs a vision language model for 2.5 seconds per image, across 10,000 images, uses 25,000 execution seconds, or 50 credits.

Volume discounts

Volume discounts are graduated: each rate applies only to the images that fall inside its band, and images below a threshold keep the rate of the band they fall in. Bands count a workspace total across all models for the calendar month. Select a band to see the rates that apply to images in it.

First 250K

Standard rates, charged on the first 250,000 images each month.

Standard rate	Models on this rate	Rate in this band
0.0625	Classification models	0.0625
0.125	Nano detection, CLIP, PP-OCR recognition only	0.125
0.1875	Small–Large detection, Nano segmentation	0.1875
0.25	XL detection, Small–Large segmentation	0.25
0.3125	XL segmentation, SAM 2 Tiny/Small	0.3125
0.375	OCR pipelines	0.375
0.5	Segment Anything	0.5
0.75	Open-vocabulary detection, Depth Anything V3 Base	0.75

250K–500K

25% off.

Standard rate	Models on this rate	Rate in this band
0.0625	Classification models	0.0469
0.125	Nano detection, CLIP, PP-OCR recognition only	0.0938
0.1875	Small–Large detection, Nano segmentation	0.1406
0.25	XL detection, Small–Large segmentation	0.1875
0.3125	XL segmentation, SAM 2 Tiny/Small	0.2344
0.375	OCR pipelines	0.2813
0.5	Segment Anything	0.375
0.75	Open-vocabulary detection, Depth Anything V3 Base	0.5625

500K–1M

50% off.

Standard rate	Models on this rate	Rate in this band
0.0625	Classification models	0.0313
0.125	Nano detection, CLIP, PP-OCR recognition only	0.0625
0.1875	Small–Large detection, Nano segmentation	0.0938
0.25	XL detection, Small–Large segmentation	0.125
0.3125	XL segmentation, SAM 2 Tiny/Small	0.1563
0.375	OCR pipelines	0.1875
0.5	Segment Anything	0.25
0.75	Open-vocabulary detection, Depth Anything V3 Base	0.375

1M–2M

60% off.

Standard rate	Models on this rate	Rate in this band
0.0625	Classification models	0.025
0.125	Nano detection, CLIP, PP-OCR recognition only	0.05
0.1875	Small–Large detection, Nano segmentation	0.075
0.25	XL detection, Small–Large segmentation	0.1
0.3125	XL segmentation, SAM 2 Tiny/Small	0.125
0.375	OCR pipelines	0.15
0.5	Segment Anything	0.2
0.75	Open-vocabulary detection, Depth Anything V3 Base	0.3

2M–5M

70% off.

Standard rate	Models on this rate	Rate in this band
0.0625	Classification models	0.0188
0.125	Nano detection, CLIP, PP-OCR recognition only	0.0375
0.1875	Small–Large detection, Nano segmentation	0.0563
0.25	XL detection, Small–Large segmentation	0.075
0.3125	XL segmentation, SAM 2 Tiny/Small	0.0938
0.375	OCR pipelines	0.1125
0.5	Segment Anything	0.15
0.75	Open-vocabulary detection, Depth Anything V3 Base	0.225

Over 5M

80% off.

Standard rate	Models on this rate	Rate in this band
0.0625	Classification models	0.0125
0.125	Nano detection, CLIP, PP-OCR recognition only	0.025
0.1875	Small–Large detection, Nano segmentation	0.0375
0.25	XL detection, Small–Large segmentation	0.05
0.3125	XL segmentation, SAM 2 Tiny/Small	0.0625
0.375	OCR pipelines	0.075
0.5	Segment Anything	0.1
0.75	Open-vocabulary detection, Depth Anything V3 Base	0.15

For example, 600,000 images through RF-DETR Small in one month bill as 250,000 at 0.1875, then 250,000 at 0.1406, then 100,000 at 0.0938, for a total of 91.41 credits.

Models priced elsewhere

The video trackers SAM 2 Video and SAM 3 Video run on Serverless Video and are priced in Serverless Video Streaming.