Model Family
OpenAI
OpenAI develops multimodal AI systems across ChatGPT and the OpenAI API. Its image-generation path has moved from the DALL-E family toward GPT Image models, which support natural-language generation, image editing, and configurable output settings through OpenAI's image tools and API surfaces.
2 models · 773 keywords · 9260 images on blupromptsGPT Image 2773 keywords · 9127 images
Model Snapshot
GPT Image 2 is the main OpenAI comparison model in the visual dictionary. It supports natural-language image generation and editing, with configurable output settings such as size, quality, format, and compression, making it the clearest current contrast to the older DALL-E 3 results.
- Best For
- detailed prompt interpretation, image editing, flexible output settings, high-fidelity handling of reference/input images, and current OpenAI comparisons
- Pricing Note
- GPT Image 2 uses token-based pricing with separate rates for image input, image output, text input, cached input, and batch usage. OpenAI also publishes estimated request costs for common output sizes and quality levels, so exact pricing can be included if the page is maintained against OpenAI's current pricing docs.
- Release Context
- OpenAI's image generation guide names GPT Image models, including `gpt-image-2`, and shows `gpt-image-2` directly in Images API examples. Because OpenAI frames GPT Image as a model family, the safest wording is that GPT Image 2 is the current OpenAI image model shown in the published examples.
*As of June 27, 2026: OpenAI lists GPT Image 2 pricing at $8.00 per 1M image input tokens, $2.00 per 1M cached image input tokens, $30.00 per 1M image output tokens, $5.00 per 1M text input tokens, and $1.25 per 1M cached text input tokens. OpenAI's image guide estimates GPT Image 2 request costs at 1024x1024: low $0.006, medium $0.053, high $0.211; 1024x1536: low $0.005, medium $0.041, high $0.165; and 1536x1024: low $0.005, medium $0.041, high $0.165.
Source: OpenAI image generation guide
DALL-E 350 keywords · 133 images
Model Snapshot
DALL-E 3 is the legacy OpenAI baseline in the published examples. It remains useful because it captures OpenAI's prompt-adherence era before the GPT Image model line became the main OpenAI comparison point.
- Pricing Note
- DALL-E 3 belonged to an earlier OpenAI image-pricing structure, while GPT Image models use token-based pricing and image-cost estimates. Because DALL-E 3 is unavailable for new comparison work, this section should avoid active-use pricing language.
- Release Context
- OpenAI introduced DALL-E 3 as a major prompt-adherence improvement over DALL-E 2 and made it available through ChatGPT and the API. OpenAI positioned DALL-E 3 around detailed prompt following and prompt refinement through ChatGPT.
- Deprecation / Legacy Note
- DALL-E 3 is preserved as a legacy OpenAI comparison model while GPT Image models carry the current image API direction.
Source: OpenAI DALL-E 3 overview