AI models
Retrieval-Augmented Generator
Parameters
626M
Published
May 22, 2020
Retrieval-Augmented Generator is an AI model developed by Facebook, New York University (NYU) and University College London (UCL) (United States and United Kingdom), first published in May 2020. It works in the language domain, on tasks such as question answering and retrieval-augmented generation.
Epoch AI has no training-compute estimate for this model. The model has 626,000,000 parameters. It was trained on roughly 3.1M datapoints. Training ran on NVIDIA Tesla V100 PCIe 32 GB.
Access: Open weights (unrestricted). Its weights are openly available. The reference paper has 13,907 citations. Epoch AI rates the confidence of this record as confident.
Full record
- Organization
- Facebook, New York University (NYU), University College London (UCL)
- Country of organization
- United States, United Kingdom
- Domain
- Language
- Task
- Question answering, Retrieval-augmented generation
- Parameters
- 626,000,000
- Dataset size
- 3.1M
- Training hardware
- NVIDIA Tesla V100 PCIe 32 GB
- Numerical format
- FP16
- Model accessibility
- Open weights (unrestricted)
- Open weights
- Yes
- Citations
- 13,907
- Epoch confidence
- Confident
SourceEpoch AI, 'AI Models'. Published online at epoch.ai. Retrieved 2026-07-29 from https://epoch.ai/data/ai-models. Licensed under CC BY 4.0.
← All ai models