dall-e-3
by openai
Pricing
Input
$0.00 / 1M tokens
Output
$0.02 / 1M tokens
DALL·E 3 Model Suite by OpenAI
Overview
DALL·E 3 is OpenAI’s third-generation text-to-image model, released in October 2023. It builds on the architecture of DALL·E 2 with dramatic improvements in prompt fidelity, realism, and ease of use—especially when used inside ChatGPT. With tight integration into ChatGPT (Plus and Enterprise), DALL·E 3 can generate high-resolution images from simple or complex prompts, and even perform conversational image editing.
Features
- Deep ChatGPT integration with automatic prompt optimization
- High-resolution image generation (1024×1024)
- Iterative editing with conversational instructions
- Stronger content safety filters and bias mitigations
Use Cases
- Marketing design, campaigns, and branding content
- Visual storytelling, storyboarding, and concept art
- Educational illustrations and scientific diagrams
- Image variations, reimagining, and contextual edits
Limitations
- Only 1 image per generation in ChatGPT (API supports more)
- Inpainting/outpainting UI not publicly available (yet)
- Declines requests involving public figures or unsafe topics
Performance Comparison
| Model | Resolution | Prompt Fidelity | Editing Support | Best Use |
|---|---|---|---|---|
| DALL·E 2 | 1024×1024 | Moderate | Inpainting, Outpainting | Image editing, research |
| DALL·E 3 | 1024×1024+ | Very High | Conversational Editing | Creativity, illustration, prototyping |
Access & Licensing
- ChatGPT: Available via Plus and Enterprise plans, with limited free-tier usage (2 images/day as of Aug 2024)
- API: Available through OpenAI's image generation endpoint
- Microsoft tools: Integrated into Bing Image Creator and Microsoft Designer
- License: Free for personal and commercial use, subject to OpenAI policies