DETR

Training compute

4×10²⁰ FLOP

Parameters

60M

Published

May 26, 2020

DETR is an AI model developed by Facebook (United States), first published in May 2020. It works in the vision domain, on tasks such as object detection.

Training it took an estimated 4×10²⁰ FLOP of compute (estimation method: hardware). The model has 60,000,000 parameters. It was trained on roughly 826K datapoints. Training ran on NVIDIA V100. The compute alone is estimated at $960 in 2023 dollars.

Access: Open weights (unrestricted). Its weights are openly available. The reference paper has 18,148 citations. Epoch AI rates the confidence of this record as confident.

Full record

Organization: Facebook
Country of organization: United States
Domain: Vision
Task: Object detection
Training compute: 4×10²⁰ FLOP
Compute estimation method: Hardware
Parameters: 60,000,000
Dataset size: 826K
Training hardware: NVIDIA V100
Training cost (2023 USD): $960
Model accessibility: Open weights (unrestricted)
Open weights: Yes
Citations: 18,148
Epoch confidence: Confident