DALL·E 2 
DALL·E 2 is OpenAI’s earlier text-to-image system for creating and editing realistic images and art from natural-language descriptions; the official page now directs users to current ChatGPT image creation. Users should use authorized prompts and references, avoid non-consensual likenesses, inspect artifacts, verify rights and disclosure needs, protect uploads and retain human creative control.
Starting price Free + ChatGPT/API pricing
Tool Information
DALL·E 2 is OpenAI’s earlier text-to-image system for creating and editing realistic images and art from natural-language descriptions; the official page now directs users to current ChatGPT image creation. Users should use authorized prompts and references, avoid non-consensual likenesses, inspect artifacts, verify rights and disclosure needs, protect uploads and retain human creative control.
Use authorized, minimal inputs and grant integrations the least access required. Configure privacy, retention, sharing, quality, accessibility, disclosure, export, governance, moderation and spending controls. Test representative cases, verify generated facts, calculations, citations, code, financial data, voice and visual details, preserve originals and version history, and retain accountable human approval before publication, investment, customer contact, deployment or operational action.
The historical model page is free to view and current image creation is available through eligible free or paid ChatGPT access and APIs where supported. Image limits, models, resolution, API usage, subscription, taxes and current OpenAI terms vary.
AI output can be inaccurate, biased, derivative, insecure, incomplete or misleading. Review consent, copyright, likeness, voice and commercial rights, training and retention terms, renewals, refunds, platform rules and applicable law. Financial, voice, customer-service, workplace, biometric, retail and geospatial workflows require qualified human review.
F.A.Q (3)
Pros and Cons
Pros
- Creates original images and art from text descriptions
- Can combine concepts; attributes; and styles
- Historically supported image generation
- Historically supported inpainting for editing selected regions
- Historically supported outpainting beyond image boundaries
- Historically supported image variations
- Produced four times the resolution of the original DALL·E
- Was preferred over DALL·E 1 for caption matching and photorealism in OpenAI evaluations
- Included filters against violent; hateful; and adult content
- Helped establish accessible text-to-image workflows
- Its research paper and launch materials remain useful educational references
- Can serve as a historical baseline for evaluating newer image models
- Supported 256 by 256; 512 by 512; and 1024 by 1024 output sizes
- Allowed multiple images in one historical generation request
- Accepted PNG source images and masks for targeted edits
- Enabled developers to automate image variations through the Images API
Cons
- DALL·E 2 is deprecated and has been removed from the API
- OpenAI recommends GPT-Image-2 instead
- It is no longer a viable choice for a new production integration
- Image quality and prompt adherence trail newer models
- Complex composition and precise spatial instructions can fail
- Generated faces; hands; and text can contain artifacts
- Safety filters may block legitimate prompts
- The model can reproduce demographic and cultural biases
- Photorealistic output creates misinformation and impersonation risk
- Rights to prompts; uploads; trademarks; and likenesses still require review
- Historical documentation may imply availability that no longer exists
- API rate and access information is now relevant mainly to legacy records
- Outputs should not be treated as factual evidence
- The original product required human monitoring to guard against misuse
- Rebuilding a workflow around a removed endpoint requires migration work
- Historical editing inputs had strict square-PNG and file-size requirements
Reviews
You must be logged in to submit a review.
No reviews yet. Be the first to review!
Before you subscribe
- Check the currency, billing period, automatic renewal, usage limits and seats at checkout — prices and promotions change.
- App-store pricing can differ from website pricing by country, device and tax.
- Share only authorised data, and read the provider's encryption, retention and model-training terms first.
- AI output can be incomplete or wrong. Verify facts, citations, calculations and code before you publish or ship.
- Commercial usage rights vary by plan, provider terms and jurisdiction.
- None of these tools replaces a qualified professional for medical, legal or financial decisions.
Promote
Share this Tool
Similar Tools
DaVinciFace uses a generative neural network to transform a submitted human portrait into an artwork inspired by Leonardo da Vinci’s visual style. Users should upload authorized photographs, obtain subject consent, protect facial data, inspect resemblance and artifacts, review deletion and publication settings, verify personal or commercial rights and retain human creative judgment.
August 20, 2026
Paid per portrait
Deep Art Effects provides web and desktop tools for transforming photos, generating images and video, and integrating creative imaging into business workflows, including offline processing options. Creators should use authorized assets, inspect style similarity and generated detail, protect private images, verify local and cloud behavior, review output licenses, preserve originals and retain human creative approval.
August 20, 2026
Free + from $9.90/mo
Deep Image is an AI image-enhancement platform for upscaling, sharpening, denoising, background processing, generative detail, avatars and API-based bulk workflows. Users should process authorized images, preserve originals, inspect invented detail and facial changes, verify output dimensions and color, protect client files, review retention and licensing and retain human quality approval.
August 20, 2026
Free + from $7.50/mo