Robots Atlas>ROBOTS ATLAS
DALL-E 3

DALL-E 3

DALL·E 3 · Family: DALL-E
OpenAI’s text-to-image model (2023) integrated with ChatGPT: better prompt following and text rendering. Now a previous generation (replaced by GPT Image).
⚠ Deprecated✓ Public accessImage generation📁 DALL-E
Release date
1 October 2023
Access:APIHostedDeployment:☁ Cloud

Overview

DALL·E 3 is a text-to-image generation model developed by OpenAI, announced in September 2023 and released in ChatGPT (for Plus and Enterprise users) in October 2023, and via the API in early November 2023. Compared with DALL·E 2 it understands nuance and detail far better and follows complex prompts more accurately.

The model is natively integrated with ChatGPT — the assistant helps expand and refine the prompts passed to the generator. DALL·E 3 also markedly improved the rendering of text within images relative to earlier generations.

It takes a natural-language description as input and produces an image as output. Via the API it supports 1024×1024, 1024×1536 and 1536×1024 resolutions and two quality levels: standard and HD. Per-image pricing ranges from USD 0.04 (standard 1024×1024) to USD 0.12 (HD, rectangular formats).

DALL·E 3 is available through the OpenAI API, ChatGPT and Microsoft (Bing Image Creator / Copilot, Azure). Since February 2024 generated images include C2PA-standard metadata; the model applies safety filters (including blocking the style of living artists). In March 2025 DALL·E 3 was replaced in ChatGPT by GPT Image’s native image generation — it is now a previous-generation model.

Classification
Image generation
Family: DALL-E
Access & deployment
APIHosted
Cloud
Weights: Closed
Key parameters
📥 Input: text

Technical specification

License
Proprietary
Modalities
⬇ Input
text
⬆ Output
image

Capabilities and applications

Native model capabilities
Text-to-image generation
Generating an image from a text description (prompt). The model interprets a natural-language instruction and produces a new, coherent visual from scratch — without any input image.
Category: vision
Text rendering in images
Generating images containing legible, correctly spelled text — infographics, posters, menu cards, QR codes, captions in a specific graphic style. A key capability that separates new-generation models from early image generators.
Category: vision
Instruction following
Precisely following instructions contained in the prompt: response format, length, style, constraints (e.g. 'reply in six words'). GPT-5.1 significantly improved this capability compared to GPT-5.
Category: language

Pricing

Technical architecture

Deployment and security

🔒 Security / Enterprise
✓ Verified enterprise information
Updated: 24 Jul 2026↗ Security documentation