AI models
NÜWA
Training compute
7.2×10²¹ FLOP
Parameters
870M
Published
Nov 24, 2021
NÜWA is an AI model developed by Microsoft Research and Peking University (United States and China), first published in November 2021. It works in the multimodal, vision, image generation and 2 more domain, on tasks such as image generation, video generation, text-to-image and text-to-video.
Training it took an estimated 7.2×10²¹ FLOP of compute (estimation method: hardware). The model has 870,000,000 parameters. It was trained on roughly 5.5B datapoints. Training ran on NVIDIA A100.
Access: Unreleased. Its weights are not openly released. The reference paper has 340 citations. Epoch AI rates the confidence of this record as confident.
Full record
- Organization
- Microsoft Research, Peking University
- Country of organization
- United States, China
- Domain
- Multimodal, Vision, Image generation, Video, Language
- Task
- Image generation, Video generation, Text-to-image, Text-to-video
- Training compute
- 7.2×10²¹ FLOP
- Compute estimation method
- Hardware
- Parameters
- 870,000,000
- Dataset size
- 5.5B
- Training hardware
- NVIDIA A100
- Model accessibility
- Unreleased
- Open weights
- No
- Citations
- 340
- 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