AI models
AR-LDM
Training compute
5.1×10²⁰ FLOP
Parameters
1.5B
Published
Nov 20, 2022
AR-LDM is an AI model developed by Alibaba, University of Waterloo and Vector Institute (China and Canada), first published in November 2022. It works in the image generation domain, on tasks such as text-to-image.
Training it took an estimated 5.1×10²⁰ FLOP of compute (estimation method: hardware). The model has 1,500,000,000 parameters. Training ran on NVIDIA A100 for about 194 hours. The compute alone is estimated at $746 in 2023 dollars.
Access: Unreleased. Its weights are not openly released. It is built on top of Stable Diffusion (LDM-KL-8-G). The reference paper has 86 citations. Epoch AI rates the confidence of this record as confident.
Full record
- Organization
- Alibaba, University of Waterloo, Vector Institute
- Country of organization
- China, Canada
- Domain
- Image generation
- Task
- Text-to-image
- Training compute
- 5.1×10²⁰ FLOP
- Compute estimation method
- Hardware
- Parameters
- 1,500,000,000
- Training hardware
- NVIDIA A100
- Training time
- 194 h
- Training cost (2023 USD)
- $746
- Model accessibility
- Unreleased
- Open weights
- No
- Base model
- Stable Diffusion (LDM-KL-8-G)
- Citations
- 86
- 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