Nano Banana
GPT Image 1 — OpenAI's First Native Image Model in the API
GPT Image 1, released to the API by OpenAI on April 23, 2025, is the natively multimodal model that powered ChatGPT's viral image generation. Design tools like Figma and Photoroom adopted it for its prompt following, text rendering and world knowledge.
- OpenAI gpt-image-1
- Up to 15 reference images
- Quality: low, medium, high
- Used by Figma, Photoroom and more
More models
- Nano Banana 2.1GoogleText to image · Image to image · Up to 4K
- Nano Banana 2GoogleText to image · Image to image · Up to 4K
- Nano Banana 2 LiteGoogleText to image · Image to image · Up to 1K
- Nano Banana ProGoogleText to image · Image to image · Up to 4K
- GPT Image 2.5 SunburstOpenAIText to image · Image to image · Up to 4K
- GPT Image 2.5 FlareOpenAIText to image · Image to image · Up to 4K
- GPT Image 2OpenAIText to image · Image to image · Up to 4K
- GPT Image 1OpenAIText to image · Image to image · Up to 1K
- Seedream 5.0 ProByteDanceText to image · Image to image · Up to 2K
- Seedream 5.0 LiteByteDanceText to image · Image to image · Up to 4K
- Seedream 4.5ByteDanceText to image · Image to image · Up to 4K
- Grok Imagine Image 2.0xAIText to image · Image to image · Up to 2K
- Grok Imagine ImagexAIText to image · Image to image · Up to 1K
- Qwen Image 3.0 ProAlibabaText to image · Up to 2K
- Qwen Image 2.1AlibabaText to image · Image to image · Up to 2K
- Wan 2.7 Image ProAlibabaText to image · Up to 4K
- Wan 2.7 ImageAlibabaText to image · Up to 2K
- Z-Image TurboAlibabaText to image · Up to 2K
- FLUX.2 MaxBlack Forest LabsText to image · Up to 1MP
- FLUX.2 ProBlack Forest LabsText to image · Image to image · Up to 1MP
- FLUX.2 FlexBlack Forest LabsText to image · Up to 1MP
- FLUX Kontext ProBlack Forest LabsText to image · Image to image · Up to 1MP
- FLUX1.1 [pro] UltraBlack Forest LabsText to image · Up to 2K
- FLUX1.1 [pro]Black Forest LabsText to image · Up to 1MP
- FLUX Fast SchnellProdiaText to image · Up to 1K
- Muse Image 1.0MetaText to image · Up to 1K
- Recraft V3RecraftText to image · Up to 1K
- Reve 2.1ReveText to image · Image to image · Up to native
Why GPT Image 1 Is a Great AI Image Generator
GPT Image 1 is a dependable, well-known model for illustrations, stickers and edits with lots of references, in square and 2:3 / 3:2 formats.
Proven
The model behind ChatGPT's 2025 image boom.
Many references
Combine up to 15 images.
Quality control
Pick low, medium or high to trade speed for detail.
World knowledge
Understands real objects, styles and context.
What GPT Image 1 can do
Text to image
Illustrations, stickers and photos.
Reference edits
Merge products, people and styles.
Three formats
1:1, 2:3 and 3:2.
1K output
Choose 1K in 3 aspect ratios: 1:1, 2:3 and 3:2.
Reference inputs
Besides a text prompt you can add up to 15 reference images.
Up to 4 images per run
Generate 1–4 variations at once and keep the best; each image is billed and refunded on its own.
Use cases
Starting points you can copy into the generator above.
| Use case | Good for | Example prompt |
|---|---|---|
| Stickers | Clean cut-out art. | A sticker sheet of six cute cactus characters with different expressions, white borders |
| Style transfer | Restyle a photo. | Turn this pet photo into a soft watercolor illustration, keep the pose |
| Product mockups | Merge references. | Put the logo from image 2 on the tote bag in image 1, natural fabric texture |
| Book illustrations | Story scenes. | A little girl and a giant friendly bear walking through a snowy forest at dusk, storybook style |
GPT Image 1 in 3 steps
- 1
Write a prompt or add references
Describe subject, style, light and any text in quotes. To edit, upload reference images and say exactly what should change.
- 2
Pick size, aspect ratio and count
Choose the resolution, aspect ratio and 1–4 images. The credit cost is shown before you submit.
- 3
Generate and keep your results
Images appear in the generator and are saved to My works. Anything that fails is refunded automatically.
X Posts About GPT Image 1 — OpenAI's First Native Image Model in the API
- OpenAI Developers@OpenAIDevs
Image gen is now available in the API! We’re launching gpt-image-1, making ChatGPT’s powerful image generation capabilities available to developers worldwide starting today. ✅ More accurate, high fidelity images 🎨 Diverse visual styles ✏️ Precise image editing 🌎 Rich world… pic.twitter.com/ndC3hIpSlX
View on X - OpenAI Developers@OpenAIDevs
Figma is leveraging gpt-image-1 to generate and edit images from simple prompts, enabling designers to rapidly explore ideas and iterate visually directly in Figma. pic.twitter.com/WF33AkdtVz
View on X - OpenAI Developers@OpenAIDevs
Photoroom is using gpt-image-1 to help online sellers instantly create studio-quality visuals, lifestyle scenes, and on-model shots from a single product photo. pic.twitter.com/wzD20FJSIj
View on X - AGIラボ@ctgptlb
【速報】OpenAI、ChatGPTの画像生成を「gpt-image-1」APIとして公開 画像編集にも対応し、Adobe Firefly/Figma/HeyGen などが先行導入。 料金はテキスト入力$5／画像入力$10／画像出力$40（100Mトークン）。画像は3段階の品質から選択でき、1枚あたり約 $0.02〜 $0.19。 本日より提供開始。 pic.twitter.com/rI7f6dVvHx
View on X - OpenAI Developers@OpenAIDevs
ICYMI: Image generation is now available as a Responses API tool—and includes new capabilities: 🌊 Streaming: see partial images as they load ⤴️ Multi-turn edits: high fidelity edits with cache savings 🌐 Combine with tools like MCP or web search for generations using live… https://t.co/XshjbSc76x pic.twitter.com/sO5195kh95
View on X - OpenAI Developers@OpenAIDevs
🎨 Generating or editing images at scale? Batch API now supports GPT Image models: gpt-image-1.5, chatgpt-image-latest, gpt-image-1, gpt-image-1-mini. Submit up to 50,000 generations or edits as async jobs at 50% lower cost with separate rate limits.https://t.co/9W6TCeBGw3 pic.twitter.com/KMcBIf8UxR
View on X
YouTube Videos About GPT Image 1 — OpenAI's First Native Image Model in the API
Unveiling GPT-image-1: Rising to new heights with image generation in Azure AI Foundry
Microsoft Azure · YouTubeHands-on with OpenAI's gpt-image-1 API using Python
Build at Scale · YouTubeOpenAI’s New Image Model on API… And It Blew My Mind 🤯
Prompt Engineer 48 · YouTubeOpenAI gpt-image-1 API - Build AMAZING AI Image Apps: Here's How
All About AI · YouTubeYou Won’t Believe What OpenAI’s New Image API Can Do!
AI Unleashed · YouTubeI built an AI image startup in 10 minutes using ChatGPT (image gen API)
Alex Finn · YouTube
Frequently asked questions
GPT Image 2 is newer with better text and higher resolution; GPT Image 1 remains a stable choice with up to 15 references.
Higher quality takes longer and costs more credits, with finer detail.
From 16 to 24 credits per image, depending on resolution, aspect ratio and quality. The exact cost is shown before you submit, and failed generations are refunded automatically.
1K; aspect ratios 1:1, 2:3 and 3:2.
No. GPT Image 1 is developed by OpenAI; Nano Banana is an independent service that runs it through a model provider.
Ready to Use the Original GPT Image?
The model that started it all, still dependable for 1K images and edits with many references.
- Up to 1K output
- Up to 15 reference images
- Failed outputs refunded