AI models
Meissonic
Training compute
1.2×10²¹ FLOP
Parameters
1B
Published
Mar 13, 2025
Meissonic is an AI model developed by National University of Singapore, Skywork AI, Hong Kong University of Science and Technology (HKUST), University of California (UC) Berkeley and Zhejiang University (ZJU) (Singapore, Hong Kong, United States and China), first published in March 2025. It works in the image generation domain, on tasks such as text-to-image and image generation.
Training it took an estimated 1.2×10²¹ FLOP of compute (estimation method: hardware). The model has 1,000,000,000 parameters. Training ran on NVIDIA H100 SXM5 80GB.
Access: Open weights (unrestricted). Its weights are openly available. It is built on top of CLIP ViT-H/14 - LAION-2B. Epoch AI rates the confidence of this record as confident.
Full record
- Organization
- National University of Singapore, Skywork AI, Hong Kong University of Science and Technology (HKUST), University of California (UC) Berkeley, Zhejiang University (ZJU)
- Country of organization
- Singapore, Hong Kong, United States, China
- Domain
- Image generation
- Task
- Text-to-image, Image generation
- Training compute
- 1.2×10²¹ FLOP
- Compute estimation method
- Hardware
- Parameters
- 1,000,000,000
- Training hardware
- NVIDIA H100 SXM5 80GB
- Chip-hours
- 1.2K
- Numerical format
- BF16
- Model accessibility
- Open weights (unrestricted)
- Open weights
- Yes
- Base model
- CLIP ViT-H/14 - LAION-2B
- 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