FMMformer (2-kernel fast weight + Band20)
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 California Los Angeles (UCLA),University of Utah
- Organisation type
- Academia,Academia
- Country
- United States of America
- Published
- 5 August 2021
- Authors
- Tan M. Nguyen, Vai Suliafu, Stanley J. Osher, Long Chen, Bao Wang
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language
- Task
- Language modeling, Text classification
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
- 40M
- Training data
- tokens
40M (caption to Table 2)
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
- 2.5 × 10¹⁶ FLOP
- How it was established
- Operation counting
6 FLOP / token / parameter * 40 * 10^6 parameters * 103*10^6 tokens * 1 epoch [assumed, not given] = 2.472e+16 FLOP ________ estimation from the Algorithmic progress paper: 4.3*10^17 FLOP
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 RTX 3090 Ti
- Chips used
- 4
- Power draw
- 3.6 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.
- Record confidence
- Speculative
- Citations
- 38
- Benchmark data
- FMMformer (2-kernel fast weight + Band20)
Sources
Where this record came from and when it was last checked.
- Reference
- FMMformer: Efficient and Flexible Transformer via Decomposed Near-field and Far-field Attention
- Last updated
- 11 February 2026
What the numbers mean
Where it came from
FMMformer (2-kernel fast weight + Band20) was published by University of California Los Angeles (UCLA),University of Utah, in the country recorded as United States of America, during August 2021. It comes out of an organisation categorised as academia,Academia.
It works in the domain of Language, and is recorded as performing the task of language modeling, Text classification.
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 2.5 × 10¹⁶ FLOP, on hardware recorded as NVIDIA GeForce RTX 3090 Ti. That figure measures what producing the model cost, and has no bearing on how fast it answers.
Answers
FMMformer (2-kernel fast weight + Band20) — common questions
FMMformer (2-kernel fast weight + Band20)— 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.
FMMformer (2-kernel fast weight + Band20)— is it open source?
No. Its weights have not been published, so it exists only as a service controlled by its owner.
FMMformer (2-kernel fast weight + Band20)— how many parameters does it have?
It has a parameter count of 40M. 40M (caption to Table 2). 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.
FMMformer (2-kernel fast weight + Band20)— who created it?
It was published by University of California Los Angeles (UCLA),University of Utah, based in United States of America, an organisation categorised as academia,Academia.
FMMformer (2-kernel fast weight + Band20)— when was it released?
It was published in August 2021. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
FMMformer (2-kernel fast weight + Band20)— what is it used for?
It works in the domain of Language, and is recorded as handling the task of language modeling, Text classification. 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.
FMMformer (2-kernel fast weight + Band20)— how much compute was used to train it?
Training consumed around 2.5 × 10¹⁶ FLOP, on hardware recorded as NVIDIA GeForce RTX 3090 Ti. 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.
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.