CoRe

Closed weights Tsinghua University 12.4B parameters December 2023

No estimate

No hardware requirements for this model

The weights for this model have not been published, so it cannot be downloaded or run on your own hardware at any size. It is reachable only through its provider, and no graphics card changes that.

On record

Full specification

Everything on record for this model. Most of it describes how it was trained rather than how it runs — useful context for judging how much work went into it, and how it compares with models built at a different scale.

Origin

Who built this model, where, and when it was published.

Organisation
Tsinghua University
Organisation type
Academia
Country
China
Published
29 December 2023
Authors
Xinyu Zhu, Junjie Wang, Lin Zhang, Yuxiang Zhang, Ruyi Gan, Jiaxing Zhang, Yujiu Yang

What it does

The problem areas the model was built for. A model can carry several of each.

Domain
Mathematics, Language
Task
Quantitative reasoning, Language modeling/generation
Base model
GPT-J-6B

Size

How large the model is and how much data it was trained on. Parameters are the figure that decides whether it fits on a given graphics card.

Parameters
12.4B

"Since the default setting consists of two GPT-J (6B) and a DeBERTa-large (0.4B), we note our backbone as “GPT-J 12B”, which implies around 12.4 billion parameters in total. "

Training data
tokens

The training run

What it physically took to train: which chips, how many, for how long, and what that drew from the wall.

Training hardware
NVIDIA A100 SXM4 40 GB

Availability

Whether you can obtain the model and run it on your own hardware, which is what decides if any of the graphics-card figures on this page apply.

Weights
Closed — provider access only
Model access
Unreleased
Training code
Open (non-commercial)

no clear license https://github.com/TianHongZXY/CoRe

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Why it is tracked
SOTA improvement

We evaluate our CoRe framework on several mathematical reasoning datasets and achieve decent improvement over state-of-the-art methods, up to 9.6% increase over best baselines.

Record confidence
Speculative
Citations
109

Sources

Where this record came from and when it was last checked.

Reference
Solving Math Word Problems via Cooperative Reasoning induced Language Models
Last updated
25 May 2026

What the numbers mean

Background

CoRe was published by Tsinghua University, in China, in December 2023. The organisation is categorised as academia.

It works in Mathematics, Language, and is recorded as doing quantitative reasoning, Language modeling/generation.

It builds on GPT-J-6B, which is why it shares that model's general shape and size.

Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.

How it was trained

The reason it appears in this catalogue at all is sOTA improvement.

Answers

CoRe — common questions

01

How many parameters does CoRe have?

CoRe has 12.4B parameters. "Since the default setting consists of two GPT-J (6B) and a DeBERTa-large (0.4B), we note our backbone as “GPT-J 12B”, which implies around 12.4 billion parameters in total. ". That figure is the total, and it is what decides how much memory the model needs — roughly half a gigabyte per billion at the compression most people use.

02

Who created CoRe?

CoRe was published by Tsinghua University, based in China, categorised as academia.

03

When was CoRe released?

CoRe was published in December 2023. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

04

What is CoRe used for?

CoRe works in Mathematics, Language, and is recorded as handling quantitative reasoning, Language modeling/generation. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

05

What GPU do I need to run CoRe?

None. CoRe is a closed model — its weights were never published, so it cannot be downloaded or run on your own hardware at any price. It is reachable only through its provider.

06

Is CoRe open source?

No. CoRe has not had its weights published, so it exists only as a service controlled by its owner.

Source

Original publication

Record last updated 25 May 2026

The other direction

Looking at it from the other side?

This page starts from the model. If you already own a card and want to know everything it will run, start from the hardware instead.