Base LM + kNN LM + Continuous Cache
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
- Stanford University,Facebook AI Research
- Organisation type
- Academia,Industry
- Country
- United States of America, France
- Published
- 1 November 2019
- Authors
- Urvashi Khandelwal, Omer Levy, Dan Jurafsky, Luke Zettlemoyer, Mike Lewis
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language
- Task
- 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
- 247M
- Training data
- 103,000,000 tokens
- Epochs
- 200
"we take the exact architecture and optimization described by Baevski & Auli (2019) and use it to create a kNN-LM for inference. This model consists of 16 layers, each with 16 self-attention heads, 1024 dimensional hidden states, and 4096 dimensional feedforward layers, amounting to 247M trainable parameters."
" During this forward pass, each target token is provided a minimum of 1536 tokens of prior context for WIKITEXT-103" 200 epochs - figure 8
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
- 3.1 × 10¹⁹ FLOP
- How it was established
- Operation counting
6 FLOP / parameter / token * 247*10^6 parameters * 103000000 tokens * 200 epochs = 3.05292e+19 FLOP __________ for the Algorithmic progress paper 7.3 × 10^18 FLOP was estimated similar to supposedly base model (transformer)
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
Training code, MIT: https://github.com/urvashik/knnlm
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
- Likely
- Citations
- 1,049
- Benchmark data
- Base LM + kNN LM + Continuous Cache
"GNN-LM achieves a new state-of-the-art perplexity of 14.8 on WikiText-103"
Sources
Where this record came from and when it was last checked.
- Reference
- Generalization through Memorization: Nearest Neighbor Language Models
- Last updated
- 25 May 2026
What the numbers mean
What this model is
Base LM + kNN LM + Continuous Cache was published by Stanford University,Facebook AI Research, in the country recorded as United States of America, during November 2019. It comes out of an organisation categorised as academia,Industry.
It works in the domain of Language, and is recorded as performing the task of language modeling.
This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.
What went into building it
Training it took a computation budget of roughly 3.1 × 10¹⁹ FLOP. That figure measures what producing the model cost, and has no bearing on how fast it answers.
It was trained on a corpus of about 103,000,000 tokens of text.
The reason it appears in this catalogue at all: sOTA improvement.
Answers
Base LM + kNN LM + Continuous Cache — common questions
Base LM + kNN LM + Continuous Cache— who created it?
It was published by Stanford University,Facebook AI Research, based in United States of America, an organisation categorised as academia,Industry.
Base LM + kNN LM + Continuous Cache— when was it released?
It was published in November 2019. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
Base LM + kNN LM + Continuous Cache— 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.
Base LM + kNN LM + Continuous Cache— how much compute was used to train it?
Training consumed around 3.1 × 10¹⁹ FLOP. 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.
Base LM + kNN LM + Continuous Cache— 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.
Base LM + kNN LM + Continuous Cache— is it open source?
No. Its weights have not been published, so it exists only as a service controlled by its owner.
Base LM + kNN LM + Continuous Cache— how many parameters does it have?
It has a parameter count of 247M. "we take the exact architecture and optimization described by Baevski & Auli (2019) and use it to create a kNN-LM for inference. This model consists of 16 layers, each with 16 self-attention heads, 1024 dimensional hidden states, and 4096 dimensional feedforward layers, amounting to 247M trainable 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.
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.