FLM-101B
FLM-101B is an AI model developed by Chinese Academy of Sciences, Harbin Institute of Technology, Nanyang Technological University and Beijing Academy of Artificial Intelligence / BAAI (China and Singapore), first published in September 2023. It works in the language domain, on tasks such as language modeling/generation, question answering and chat.
Training it took an estimated 5.7×10²² FLOP of compute (estimation method: hardware). The model has 101,000,000,000 parameters. It was trained on roughly 311.5B datapoints. Training ran on NVIDIA A800 PCIe 40 GB for about 517 hours.
Access: Open weights (unrestricted). Its weights are openly available. The reference paper has 28 citations. Epoch AI rates the confidence of this record as confident.
- Organization
- Chinese Academy of Sciences, Harbin Institute of Technology, Nanyang Technological University, Beijing Academy of Artificial Intelligence / BAAI
- Country of organization
- China, Singapore
- Domain
- Language
- Task
- Language modeling/generation, Question answering, Chat
- Training compute
- 5.7×10²² FLOP
- Compute estimation method
- Hardware
- Parameters
- 101,000,000,000
- Dataset size
- 311.5B
- Training hardware
- NVIDIA A800 PCIe 40 GB
- Training time
- 517 h
- Model accessibility
- Open weights (unrestricted)
- Open weights
- Yes
- Citations
- 28
- Epoch confidence
- Confident