CoRe

Parameters

12.4B

Published

Dec 29, 2023

CoRe is an AI model developed by Tsinghua University (China), first published in December 2023. It works in the mathematics and language domain, on tasks such as quantitative reasoning and language modeling/generation.

Epoch AI has no training-compute estimate for this model. The model has 12,400,000,000 parameters. Training ran on NVIDIA A100 SXM4 40 GB.

Access: Unreleased. Its weights are not openly released. It is built on top of GPT-J-6B. The reference paper has 109 citations. Epoch AI rates the confidence of this record as speculative.

Full record

Organization: Tsinghua University
Country of organization: China
Domain: Mathematics, Language
Task: Quantitative reasoning, Language modeling/generation
Parameters: 12,400,000,000
Training hardware: NVIDIA A100 SXM4 40 GB
Model accessibility: Unreleased
Open weights: No
Base model: GPT-J-6B
Citations: 109
Epoch confidence: Speculative