BASIC-L
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
- Organisation type
- Industry
- Country
- United States of America
- Published
- 19 November 2021
- Authors
- Hieu Pham, Zihang Dai, Golnaz Ghiasi, Kenji Kawaguchi, Hanxiao Liu, Adams Wei Yu, Jiahui Yu, Yi-Ting Chen, Minh-Thang Luong, Yonghui Wu, Mingxing Tan, Quoc V. Le
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Vision
- Task
- Image classification
- Numerical format
- BF16
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
- 3.1B
- Training data
- 8,905,031,712,000 tokens
2.4B image model + 670M text model
6.7B image-text pairs
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
- 4.1 × 10²² FLOP
- How it was established
- Hardware
6.9k + 1k + 0.8k = 8.7k TPUv4 core-days for BASIC-L, per Table 8 Two cores per chip, and 275 teraflop/s per chip (https://cloud.google.com/tpu/docs/system-architecture-tpu-vm#tpu_v4) 275 teraflops * 8700/2 * 24 * 3600 * 0.4 (assumed utilization) = 8.3e22
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
- Google TPU v4
- Chip-hours
- 4,350
- Compute cost
- $1,685
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
- Likely
- Citations
- 241
SOTA on ImageNet for a model that was not trained on ImageNet images: "We present a combined scaling method – named BASIC – that achieves 85.7% top-1 accuracy on the ImageNet ILSVRC-2012 validation set without learning from any labeled ImageNet example. This accuracy surpasses best-published similar models – CLIP and ALIGN – by 9.3%"
Sources
Where this record came from and when it was last checked.
- Reference
- Combined Scaling for Zero-shot Transfer Learning
- Last updated
- 25 May 2026
What the numbers mean
Background
BASIC-L was published by Google, in the country recorded as United States of America, during November 2021. The publishing organisation is categorised as industry.
It works in the domain of Vision, and is recorded as performing the task of image classification.
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 4.1 × 10²² FLOP, on hardware recorded as Google TPU v4. 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 8,905,031,712,000 tokens of text.
The reason it appears in this catalogue at all: sOTA improvement.
Answers
BASIC-L — common questions
BASIC-L— what is it used for?
It works in the domain of Vision, and is recorded as handling the task of image classification. These are the areas it was designed around; they describe intent rather than a hard boundary.
BASIC-L— how much compute was used to train it?
Training consumed around 4.1 × 10²² FLOP, on hardware recorded as Google TPU v4. 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.
BASIC-L— 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.
BASIC-L— is it open source?
No. Its weights have not been published, so it exists only as a service controlled by its owner.
BASIC-L— how many parameters does it have?
It has a parameter count of 3.1B. 2.4B image model + 670M text model. 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.
BASIC-L— who created it?
It was published by Google, based in United States of America, an organisation categorised as industry.
BASIC-L— when was it released?
It was published in November 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.
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.