Eurus-2-7B-PRIME
Eurus-2-7B-PRIME is an AI model developed by Shanghai AI Lab, Tsinghua University, University of Illinois Urbana-Champaign (UIUC), Peking University, Shanghai Jiao Tong University and Chinese University of Hong Kong (CUHK) (China, United States and Hong Kong), first published in February 2025. It works in the language domain, on tasks such as mathematical problem solving, mathematical reasoning and language modeling/generation.
Training it took an estimated 4×10²⁰ FLOP of compute. The model has 7,000,000,000 parameters. Training ran on 8 NVIDIA A800 PCIe 80 GB.
Access: Open weights (restricted use). Its weights are openly available. It is built on top of Qwen2.5-Math-7B-Base. The reference paper has 333 citations. Epoch AI rates the confidence of this record as likely.
- Organization
- Shanghai AI Lab, Tsinghua University, University of Illinois Urbana-Champaign (UIUC), Peking University, Shanghai Jiao Tong University, Chinese University of Hong Kong (CUHK)
- Country of organization
- China, United States, Hong Kong
- Domain
- Language
- Task
- Mathematical problem solving, Mathematical reasoning, Language modeling/generation
- Training compute
- 4×10²⁰ FLOP
- Parameters
- 7,000,000,000
- Training hardware
- NVIDIA A800 PCIe 80 GB
- Chips used
- 8
- Training power draw
- 3.9 kW
- Model accessibility
- Open weights (restricted use)
- Open weights
- Yes
- Base model
- Qwen2.5-Math-7B-Base
- Citations
- 333
- Epoch confidence
- Likely