AI models
InternVL
Training compute
1.7×10²³ FLOP
Parameters
14B
Published
Jan 15, 2024
InternVL is an AI model developed by Shanghai AI Lab, Nanjing University, The University of Hong Kong, Tsinghua University, SenseTime and University of Science and Technology of China (USTC) (China and Hong Kong), first published in January 2024. It works in the vision and language domain, on tasks such as visual question answering, image classification and image captioning.
Training it took an estimated 1.7×10²³ FLOP of compute (estimation method: operation counting). The model has 14,000,000,000 parameters. Training ran on 640 NVIDIA A100 SXM4 80 GB for about 800 hours.
Access: Open weights (unrestricted). Its weights are openly available. It is built on top of InternViT-6B,LLaMA-7B. Epoch AI rates the confidence of this record as likely.
Full record
- Organization
- Shanghai AI Lab, Nanjing University, The University of Hong Kong, Tsinghua University, SenseTime, University of Science and Technology of China (USTC)
- Country of organization
- China, Hong Kong
- Domain
- Vision, Language
- Task
- Visual question answering, Image classification, Image captioning
- Training compute
- 1.7×10²³ FLOP
- Compute estimation method
- Operation counting
- Parameters
- 14,000,000,000
- Training hardware
- NVIDIA A100 SXM4 80 GB
- Chips used
- 640
- Training time
- 800 h
- Training power draw
- 507.1 kW
- Numerical format
- BF16
- Model accessibility
- Open weights (unrestricted)
- Open weights
- Yes
- Base model
- InternViT-6B, LLaMA-7B
- Epoch confidence
- Likely
More from Shanghai AI Lab,Nanjing University,The University of Hong Kong,Tsinghua University,SenseTime,University of Science and Technology of China (USTC)
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