DALL·E 2

Closed weights OpenAI 3.5B parameters April 2022

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
OpenAI
Organisation type
Industry
Country
United States of America
Published
6 April 2022
Authors
Aditya Ramesh, Prafulla Dhariwal, Alex Nichol, Casey Chu, Mark Chen

What it does

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

Domain
Image generation
Task
Text-to-image, Image generation
Approach
Self-supervised learning

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
3.5B

"Our decoder architecture is the 3.5 billion parameter GLIDE model"

Training data
167,050,000,000 tokens

"When training the encoder, we sample from the CLIP [39] and DALL-E [40] datasets (approximately 650M images in total) with equal probability"

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
3.4 × 10²³ FLOP

Decoder architecture is similar to Imagen (1.46E+22), but trained on 1.6e9 datapoints (Table 3) rather than Imagen's 5.1e9 datapoints. DALL-E 2 uses two models as priors. I estimate the prior model's FLOP as 6*N*D = 6 * 1e9 * 4096 * 1e6 = 2.5e19 FLOP. However, this seems low compared to CLIP. So it may be possible to estimate DALL-E 2's compute by analogy to Imagen, but there is a lot of uncertainty and more research would be needed. here (https://arxiv.org/pdf/2407.15811) they claim the DALL…

How it was established
Third-party estimation

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
API access
Training code
Unreleased

How it is classified

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

Foundation model
Yes
Likely above 10²³ FLOP
Yes
Why it is tracked
Highly cited,SOTA improvement

Table 2

Record confidence
Speculative
Citations
8,200

Sources

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

Reference
Hierarchical Text-Conditional Image Generation with CLIP Latents
Last updated
1 January 2026

What the numbers mean

What this model is

DALL·E 2 was published by OpenAI, in United States of America, in April 2022. industry is the category the publisher falls under.

It works in Image generation, and is recorded as doing text-to-image, Image generation.

Its weights were never published, so it can only be reached through its provider. No graphics card changes that.

How it was trained

Producing it required around 3.4 × 10²³ FLOP of arithmetic, which is a statement about the training budget rather than about inference.

It was trained on about 167,050,000,000 tokens of text.

Its inclusion criterion is highly cited,SOTA improvement.

Answers

DALL·E 2 — common questions

01

How much compute was used to train DALL·E 2?

Around 3.4 × 10²³ 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.

02

What GPU do I need to run DALL·E 2?

None. DALL·E 2 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.

03

Is DALL·E 2 open source?

No. DALL·E 2 has not had its weights published, so it exists only as a service controlled by its owner.

04

How many parameters does DALL·E 2 have?

DALL·E 2 has 3.5B parameters. "Our decoder architecture is the 3.5 billion parameter GLIDE 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.

05

Who created DALL·E 2?

DALL·E 2 was published by OpenAI, based in United States of America, categorised as industry.

06

When was DALL·E 2 released?

DALL·E 2 was published in April 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.

07

What is DALL·E 2 used for?

DALL·E 2 works in Image generation, and is recorded as handling text-to-image, Image generation. These are the areas it was designed around; they describe intent rather than a hard boundary.

Source

Original publication

Record last updated 1 January 2026

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.