retrieval-quality-kNN-LMs
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
- University of Massachusetts Amherst
- Organisation type
- Academia
- Country
- United States of America
- Published
- 28 October 2022
- Authors
- Andrew Drozdov, Shufan Wang, Razieh Rahimi, Andrew McCallum, Hamed Zamani, Mohit Iyyer
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language
- Task
- Language modeling
- Base model
- Base LM + kNN LM + Continuous Cache
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
- 247M
- Training data
- 103,000,000 tokens
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 source
MIT for code: https://github.com/iesl/knnlm-retrieval-quality
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Citations
- 15
- Benchmark data
- retrieval-quality-kNN-LMs
Sources
Where this record came from and when it was last checked.
- Reference
- You can’t pick your neighbors, or can you? When and how to rely on retrieval in the kNN-LM
- Last updated
- 28 November 2025
What the numbers mean
Background
retrieval-quality-kNN-LMs was published by University of Massachusetts Amherst, in the country recorded as United States of America, during October 2022. It comes out of an organisation categorised as academia.
It works in the domain of Language, and is recorded as performing the task of language modeling.
Its starting point was an existing base model, Base LM + kNN LM + Continuous Cache. That is why it shares the base model's general shape and size.
Its weights were never published, so it can only be reached through its provider. No graphics card changes that.
How it was trained
The training set ran to roughly 103,000,000 tokens of text.
Answers
retrieval-quality-kNN-LMs — common questions
retrieval-quality-kNN-LMs— what is it used for?
It works in the domain of Language, and is recorded as handling the task of language modeling. These are the areas it was designed around; they describe intent rather than a hard boundary.
retrieval-quality-kNN-LMs— 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.
retrieval-quality-kNN-LMs— is it open source?
No. Its weights have not been published, so it exists only as a service controlled by its owner.
retrieval-quality-kNN-LMs— how many parameters does it have?
It has a parameter count of 247M. 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.
retrieval-quality-kNN-LMs— who created it?
It was published by University of Massachusetts Amherst, based in United States of America, an organisation categorised as academia.
retrieval-quality-kNN-LMs— when was it released?
It was published in October 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.
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.