CELLE-2

Open weights Chan Zuckerberg Initiative,University of California San Francisco,University of California (UC) Berkeley October 2023

No estimate

No hardware requirements for this model

This model's weights are open, but no parameter count has been published for it. Every memory and speed figure starts from that number, so we would rather show nothing than a fabricated estimate.

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
Chan Zuckerberg Initiative,University of California San Francisco,University of California (UC) Berkeley
Organisation type
Research collective,Academia,Academia
Country
United States of America
Published
10 October 2023
Authors
Emaad Khwaja, Yun Song, Aaron Agarunov, Bo Huang

What it does

The problem areas the model was built for. A model can carry several of each.

Domain
Biology
Task
Protein localization prediction, Text-to-image

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.

Training data
4,327,372 tokens

Image tokens: 17,268 × 256 = 4,420,608 Sequence tokens: 17,268 × 400 = 6,907,200 Total: 4,420,608 + 6,907,200 = 11,327,808 (1.13e7) ['Likely' confidence in dataset size estimation - two estimations slightly differ within same OOM]

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
Open — downloadable
Model access
Open weights (unrestricted)
Training code
Open source

MIT license: https://huggingface.co/HuangLab/CELL-E_2_HPA_Finetuned_480 MIT license: https://github.com/BoHuangLab/CELL-E_2

Hugging Face
HuangLab

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Record confidence
Likely
Citations
3

Sources

Where this record came from and when it was last checked.

Reference
CELL-E 2: Translating Proteins to Pictures and Back with a Bidirectional Text-to-Image Transformer
Last updated
28 November 2025

What the numbers mean

Where it came from

CELLE-2 was published by Chan Zuckerberg Initiative,University of California San Francisco,University of California (UC) Berkeley, in United States of America, in October 2023. research collective,Academia,Academia is the category the publisher falls under.

It works in Biology, and is recorded as doing protein localization prediction, Text-to-image.

Published weights mean the model runs on your machine rather than someone else's, which is what makes the hardware question below answerable at all. It is published under the HuangLab organisation on Hugging Face.

How it was trained

It was trained on about 4,327,372 tokens of text.

Answers

CELLE-2 — common questions

01

What GPU do I need to run CELLE-2?

We cannot say. CELLE-2 has open weights, but no parameter count has been published for it, and every memory and speed calculation starts from that number. We would rather show nothing than a fabricated estimate.

02

Is CELLE-2 open source?

Its weights are published, so CELLE-2 can be downloaded and run on your own hardware. Note that open weights is not the same as open source in the full sense — it says nothing about the training data, the training code, or the commercial terms attached.

03

How many parameters does CELLE-2 have?

No parameter count has been published for CELLE-2, which is why no memory or speed figure appears on this page.

04

Who created CELLE-2?

CELLE-2 was published by Chan Zuckerberg Initiative,University of California San Francisco,University of California (UC) Berkeley, based in United States of America, categorised as research collective,Academia,Academia.

05

When was CELLE-2 released?

CELLE-2 was published in October 2023. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

06

What is CELLE-2 used for?

CELLE-2 works in Biology, and is recorded as handling protein localization prediction, Text-to-image. These are the areas it was designed around; they describe intent rather than a hard boundary.

07

Where can I download CELLE-2?

Its weights are published under the HuangLab organisation on Hugging Face. We do not host model files — this site calculates what hardware is needed to run them.

Source

Original publication

Record last updated 28 November 2025

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.