Vega v2
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
- Wuhan University,JD Explore Academy,Shanghai AI Lab,Nanyang Technological University,Washington University in St Louis,Chongqing University of Posts and Telecommunications,University of Sydney
- Organisation type
- Academia,Academia,Academia,Academia,Academia,Academia
- Country
- China, Singapore, United States of America, Australia
- Published
- 4 December 2022
- Authors
- Qihuang Zhong, Liang Ding, Yibing Zhan, Yu Qiao, Yonggang Wen, Li Shen, Juhua Liu, Baosheng Yu, Bo Du, Yixin Chen, Xinbo Gao, Chunyan Miao, Xiaoou Tang, Dacheng Tao
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language
- Task
- Language modeling, Question answering, Word sense disambiguation
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
- 6B
- Training data
- 6,439,999,999 tokens
Training compute
The arithmetic performed to train the model, measured in floating-point operations. It is a measure of what the training run cost, not of how fast the finished model answers you.
- Training compute
- 7.8 × 10²² FLOP
Pretraining took 1 month on 320 A100 (Section 3.1) 720*60*60*320*312000000000000*0.3=7.7635584e+22
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
- Chips used
- 320
- Wall-clock time
- 720 hours (30 days)
- Power draw
- 255.9 kW
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
- Unreleased
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
- Record confidence
- Confident
- Citations
- 42
Highest score at SuperGLUE leaderboard version 2.0 in terms of RTE (Recognizing Textual Entailment-Accuracy)
Sources
Where this record came from and when it was last checked.
- Reference
- Toward Efficient Language Model Pretraining and Downstream Adaptation via Self-Evolution: A Case Study on SuperGLUE
- Last updated
- 25 May 2026
What the numbers mean
Background
Vega v2 was published by Wuhan University,JD Explore Academy,Shanghai AI Lab,Nanyang Technological University,Washington University in St Louis,Chongqing University of Posts and Telecommunications,University of Sydney, in the country recorded as China, during December 2022. The category the publisher falls under is academia,Academia,Academia,Academia,Academia,Academia.
It works in the domain of Language, and is recorded as performing the task of language modeling, Question answering, Word sense disambiguation.
Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.
How it was trained
Producing it required arithmetic totalling around 7.8 × 10²² FLOP, on hardware recorded as NVIDIA A100. That figure measures what producing the model cost, and has no bearing on how fast it answers.
The training set ran to roughly 6,439,999,999 tokens of text.
Its inclusion criterion: sOTA improvement.
Answers
Vega v2 — common questions
Vega v2— is it open source?
No. Its weights have not been published, so it exists only as a service controlled by its owner.
Vega v2— how many parameters does it have?
It has a parameter count of 6B. 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.
Vega v2— who created it?
It was published by Wuhan University,JD Explore Academy,Shanghai AI Lab,Nanyang Technological University,Washington University in St Louis,Chongqing University of Posts and Telecommunications,University of Sydney, based in China, an organisation categorised as academia,Academia,Academia,Academia,Academia,Academia.
Vega v2— when was it released?
It was published in December 2022. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
Vega v2— what is it used for?
It works in the domain of Language, and is recorded as handling the task of language modeling, Question answering, Word sense disambiguation. A model can carry several of each, so these are the areas it was built for rather than a limit on what it will attempt.
Vega v2— how much compute was used to train it?
Training consumed around 7.8 × 10²² FLOP, on hardware recorded as NVIDIA A100. That measures what producing the model cost and says nothing about how quickly it answers once trained — inference speed comes from memory bandwidth, not from the training budget.
Vega v2— what GPU do I need to run it?
None. This 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.
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.