OpenAI GPT Image-1 generates images from text prompts from OpenAI's latest text-to-image model, ideal for creating visual assets. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
WEJŚCIE
WYJŚCIE
BezczynnyKażde uruchomienie będzie kosztować 0.011. Za $10 możesz uruchomić ten model około 909 razy.
Co możesz zrobić dalej:
Schema wejściowy
Następujące parametry są akceptowane w treści żądania.
Brak dostępnych parametrów.
Przykładowa treść żądania
{
"model": "openai/gpt-image-1/text-to-image"
}Zaloguj się, aby wyświetlić historię zapytań
Musisz być zalogowany, aby uzyskać dostęp do historii zapytań modelu.
Zaloguj sięOpenAI GPT Image 1
GPT Image 1 is OpenAI’s latest multimodal image generation model, built to understand both text and image inputs and produce visually coherent, high-quality image outputs. It combines the reasoning power of GPT-4-Turbo with DALL·E-class visual synthesis, allowing for creative, controllable, and context-aware generation across illustration, photography, design, and visualization tasks.
🧠 Key Features
-
Multimodal Understanding
Accepts both text and image inputs, enabling style transfer, editing, or contextual composition. -
Flexible Styles
Produces photorealistic renders, stylized artwork, concept art, infographics, and 3D-style illustrations. -
High Visual Fidelity
Maintains object relationships, lighting consistency, and color balance with strong adherence to prompts. -
Accurate Text Rendering
Capable of generating clean typography, ideal for posters, memes, comics, and branding visuals. -
Knowledge-Grounded Creativity
Uses GPT-4’s world knowledge to generate factual, contextually appropriate visuals.
⚙️ Parameters
| Parameter | Description |
|---|---|
prompt | Required text description of the desired image |
size | Supports 1024×1024, 1024×1536, and 1536×1024 |
quality | Choose between low, medium, and high |
💡 Tips for Best Results
- Write prompts that specify style, subject, composition, and lighting.
- Example:
A small robot exploring an abandoned city, cartoon style, bright colors.
- Use high quality for detailed or large-format outputs.
- Prefer landscape (
1536×1024) for cinematic or wide compositions. - Prefer portrait (
1024×1536) for characters or vertical art.
📝 Notes
- All generated content follows OpenAI’s safety and content policies.
- If a prompt triggers moderation, rephrase or simplify it.
- This model supports multi-image input via API, enabling creative editing and composition workflows.
- For performance- and latency-sensitive cases, use medium quality as the balanced default.
