ChatGPT Image Generation
imageby openai
Native image generation inside ChatGPT powered by ChatGPT Images 2.0 (gpt-image-2), with accurate text rendering, multi turn consistency, and in context learning from uploaded references.
Key features
Free tier available, Plus at $20/mo
ChatGPT users who want image generation without switching to a separate tool
Integrated directly into ChatGPT, so there is no need to switch to a separate image generation tool
What it does
ChatGPT Images 2.0 (gpt-image-2)
Images are generated natively by ChatGPT Images 2.0 (gpt-image-2), meaning the model deeply understands your prompt and conversation context. This is not a separate image model being called; it is one unified model. It ships in two modes: Instant Mode for all users and Thinking Mode (Plus and above) with reasoning, web search, and multi image consistency.
Learn moreDramatically Improved Text Rendering
Renders readable, accurate text inside images with dramatically improved quality. Logos, posters, memes, signs, diagrams, infographics, and social media graphics with text overlays are all handled reliably. A major leap beyond DALL-E 3, which frequently garbled or misspelled text.
Learn moreMulti Turn Consistency
Characters, styles, and visual elements persist across multiple turns of conversation. Ask for a character illustration, then ask for variations or new scenes with the same character, and the model maintains visual consistency.
Learn moreIn Context Learning from Uploads
Upload reference images and the model learns from them in context. Share a brand style guide, a photo of a product, or an example illustration, and the model matches the style, colors, or subject in its generations.
Learn moreConversational Image Editing
Edit generated images by describing changes in natural language. Tell ChatGPT what to modify (colors, elements, text, composition) and the model applies targeted edits while preserving the rest of the image. No need to regenerate from scratch.
Learn morePhotorealistic Quality
Photorealistic output quality has improved significantly with Images 2.0. Scenes, portraits, and product photography look more natural and detailed, with stronger lighting, textures, and spatial coherence.
Learn moreTransparent Backgrounds
Generate images with transparent backgrounds for logos, stickers, icons, and design assets that need to be composited onto other materials.
C2PA Provenance Metadata
All generated images include C2PA metadata, an open standard for content provenance that lets anyone verify the image was AI generated and trace it back to OpenAI.
DALL-E 3 Retired
DALL-E 2 and DALL-E 3 were retired on May 12, 2026. All image generation in ChatGPT now uses ChatGPT Images 2.0 (gpt-image-2). Developers who had integrations on DALL-E endpoints need to migrate to gpt-image-2.
Works with GPT-5.5 Models
Image generation is available across the latest ChatGPT model lineup, including GPT-5.5 Instant and GPT-5.5 Thinking, ensuring access to the most capable text understanding alongside image output.
Pricing
Free
Limited image generation via Images 2.0 Instant Mode. Uses GPT-5.3 Instant. US users see ads.
- Images 2.0 Instant Mode
- GPT-5.3 Instant
- Text rendering supported
- C2PA metadata included
- Ads shown (US)
Go
Expanded image generation limits beyond Free. Images 2.0 Instant Mode with more generous allowances. US users see ads.
- Expanded daily image limits
- Images 2.0 Instant Mode
- Text rendering supported
- C2PA metadata included
- Ads shown (US)
Plus
Generous image generation limits. Images 2.0 with both Instant Mode and Thinking Mode (reasoning, web search, consistency across up to 8 images). Access to GPT-5.5 Thinking.
- Images 2.0 Instant and Thinking Mode
- GPT-5.5 Thinking access
- Transparent backgrounds
- Multi turn consistency
- In context learning from uploads
Pro $100
5x Plus usage limits. Access to GPT-5.5 Pro model. Images 2.0 with Thinking Mode. Launched April 9, 2026.
- 5x Plus image generation limits
- GPT-5.5 Pro model access
- Images 2.0 Instant and Thinking Mode
- All Plus features
Pro $200
20x Plus usage limits. Unlimited image generation with the highest quality output. 1M token context window. Full access to all image generation features.
- 20x Plus image generation limits
- Unlimited image generation
- Highest quality output
- 1M token context window
- All Pro $100 features
- Non-watermarked Sora video
Pros & Cons
Pros
- Integrated directly into ChatGPT, so there is no need to switch to a separate image generation tool
- Dramatically improved text rendering produces accurate, readable text in images, a major leap over DALL-E 3 and most competing models
- Conversational image editing lets you describe changes in chat and the model applies them without regenerating from scratch
- Photorealistic quality has improved significantly with Images 2.0, and Thinking Mode adds reasoning and web search grounding for more complex compositions
- Multi turn consistency keeps characters and styles visually coherent across multiple generations in the same conversation
- In context learning from uploaded reference images lets you match brand styles, product photos, or illustration aesthetics
- Free tier is available, so anyone can try image generation without paying
Cons
- Quality can be inconsistent on complex scenes with many elements, especially beyond 10 to 20 objects
- The gpt-image-2 API is available but billed per token, and Thinking Mode adds reasoning token overhead that can make complex prompts 3 to 5x more expensive than Instant Mode
- Free tier is limited to Instant Mode only and US users see ads
- Pro plans at $100/mo and $200/mo are expensive if image generation is your primary use case
- DALL-E 2 and DALL-E 3 were retired on May 12, 2026, so any existing DALL-E integrations need migration to gpt-image-2
How to get started
Open ChatGPT
Go to chatgpt.com or open the ChatGPT desktop/mobile app. Sign in with your account. Image generation works on all tiers including Free.
Describe what you want
Type a description of the image you want to create. Be specific about the subject, style, colors, composition, and any text you want included. For example: "Create a minimalist logo for a coffee shop called 'Bean & Brew' with a coffee cup icon and earth tones."
Iterate with follow ups
Refine the result using follow up messages. The model remembers the previous generation and maintains consistency. Say things like "Make the text larger," "Change the background to white," or "Keep the same style but show it from a different angle."
Upload reference images
For style matching or brand consistency, upload a reference image and ask ChatGPT to generate something in the same style. This in context learning capability lets the model pick up on colors, textures, and artistic choices from your example.
Download and use
Click the generated image to view it at full resolution. Download it directly. All images include C2PA metadata for provenance tracking. Request transparent backgrounds when creating logos or design assets.
Deep dive
Detailed guides with comparisons, tips, and visuals for each feature.
How ChatGPT Images 2.0 Works
Why native generation inside the language model is fundamentally different from DALL-E, and what that means for quality, consistency, and instruction following.
Creative Workflows and Best Practices
How to get the best results from ChatGPT image generation, including reference image techniques, style consistency, iterative refinement, and text overlay strategies.
ChatGPT Image Gen vs Midjourney vs Nano Banana vs Stable Diffusion
How ChatGPT's native image generation compares to the leading alternatives across quality, features, pricing, and workflow.
Links
Official
Learn
Documentation
Pricing
Similar Tools
ChatGPT
openai
chatbotThe most popular AI assistant in the world: text, images, video, voice, search, coding agents, and more in one place.
Nano Banana 2
Google's fastest AI image generator. Pro quality at Flash speed, with text rendering, 4K resolution, and image editing built into Gemini.
Google Whisk
DISCONTINUED: Google Whisk was shut down on April 30, 2026. Users have been migrated to Flow.
Get notified about updates
We'll email you when this tool's pricing or features change.
Last updated: 2026-06-01