NAS+ESS (23M)
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
- Northeastern University (China),NiuTrans Research,Kingsoft
- Organisation type
- Academia,Industry
- Country
- China
- Published
- 6 May 2020
- Authors
- Yinqiao Li, Chi Hu, Yuhao Zhang, Nuo Xu, Yufan Jiang, Tong Xiao, Jingbo Zhu, Tongran Liu, Changliang Li
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language
- Task
- Neural Architecture Search - NAS, Language modeling
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
- 23M
- Training data
- tokens
- Epochs
- 3,000
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 GeForce GTX 1080 Ti
- Chips used
- 1
- Power draw
- 280 W
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
- Benchmark data
- NAS+ESS (23M)
"Our ESS method achieves state-of-the-art result on the PTB task"
Sources
Where this record came from and when it was last checked.
- Reference
- Learning Architectures from an Extended Search Space for Language Modeling
- Last updated
- 11 February 2026
What the numbers mean
About this model
NAS+ESS (23M) was published by Northeastern University (China),NiuTrans Research,Kingsoft, in China, in May 2020. academia,Industry is the category the publisher falls under.
It works in Language, and is recorded as doing neural Architecture Search - NAS, Language modeling.
This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.
How it was trained
Its inclusion criterion is sOTA improvement.
Answers
NAS+ESS (23M) — common questions
When was NAS+ESS (23M) released?
NAS+ESS (23M) was published in May 2020. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
What is NAS+ESS (23M) used for?
NAS+ESS (23M) works in Language, and is recorded as handling neural Architecture Search - NAS, Language modeling. These are the areas it was designed around; they describe intent rather than a hard boundary.
What GPU do I need to run NAS+ESS (23M)?
None. NAS+ESS (23M) 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.
Is NAS+ESS (23M) open source?
No. NAS+ESS (23M) has not had its weights published, so it exists only as a service controlled by its owner.
How many parameters does NAS+ESS (23M) have?
NAS+ESS (23M) has 23M parameters. 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.
Who created NAS+ESS (23M)?
NAS+ESS (23M) was published by Northeastern University (China),NiuTrans Research,Kingsoft, based in China, categorised as academia,Industry.
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.