HiDream O1 1.5 Edit API

HiDream O1 Image

HiDream O1 Image is a state-of-the-art image generation model by HiDream AI, supporting text-to-image, image editing, and subject-driven personalization in a single unified interface. With strong prompt fidelity, flexible aspect ratio control, and multi-reference personalization, HiDream O1 delivers high-quality creative output for professional and consumer use cases alike.

🌟 Key Features

✏️ Text-to-Image

Generate high-resolution images from detailed text prompts with strong semantic understanding and photorealistic quality.

🖼 Image Editing

Provide a single reference image to edit, transform, or stylize it while preserving the original composition and subject identity.

🎭 Subject-Driven Personalization

Supply multiple reference images to drive subject-consistent generation across varied scenes, styles, and contexts.

📐 Flexible Output Sizes

Choose from six standard presets — square HD, portrait, landscape — or match the reference image's original aspect ratio automatically.

🔁 Reproducible Results

Pin a seed value to reproduce or iterate on a specific generation, enabling systematic prompt refinement.

⚙️ Parameters

Parameter	Required	Description
prompt	✅	Text description of the image to generate. Maximum 5,000 characters.
reference_image_urls	✅	List of reference image URLs. Omit for text-to-image; one URL for image editing; multiple URLs for personalization.
image_size	❌	Output size preset: square_hd, square, portrait_4_3, portrait_16_9, landscape_4_3 (default), landscape_16_9.
num_inference_steps	❌	Denoising steps (1–100, default: 50). Higher values improve quality but increase time.
guidance_scale	❌	Prompt adherence strength (1.0–20.0, default: 5.0). Higher values follow the prompt more strictly.
output_format	❌	Output format: png (default), jpeg, or webp.

💲 Pricing

Billing Standard	Price
Per image	$0.044

💡 Prompt Tips

A lone astronaut standing on a rust-colored alien desert, vast canyon in the background, golden hour lighting, cinematic composition, photorealistic.

• Be specific about subject, environment, lighting, and camera angle for best results.
• For image editing, describe what you want changed while referencing the original subject.
• For personalization, provide 2–4 reference images from different angles or contexts.
• Use guidance_scale between 4–7 for balanced creativity; go higher (8–12) when strict prompt adherence is needed.
• Fix seed when iterating on a prompt to isolate the effect of each change.

🎯 Use Cases

• Creative Illustration — Generate concept art, characters, and environments from descriptive text prompts.
• Product Visualization — Transform product photos into styled marketing visuals with image editing mode.
• Brand Personalization — Use multi-reference personalization to generate consistent subject appearances across campaigns.
• Social Media Content — Quickly produce scroll-stopping visuals in portrait or square formats for any platform.
• Rapid Prototyping — Iterate on creative directions at scale before committing to production assets.
• Style Transfer — Apply artistic styles to existing images while preserving key compositional elements.

🔄 Generation Modes

Mode	reference_image_urls	Description
Text-to-Image	Not provided	Pure text-driven image generation
Image Editing	1 URL	Edit or stylize a single reference image
Personalization	2+ URLs	Subject-consistent generation across scenes

📝 Notes

• prompt is the only required field (besides model).
• reference_image_urls accepts publicly accessible image URLs.
• keep_original_aspect only applies when reference_image_urls contains exactly one image.
• Task status values: created, processing, completed, failed.
• Generated image URLs are returned in outputs once status is completed.