hidream-ai/hidream-o1/dev
Hidream O1 Dev by hidream-ai - generate stunning images from text prompts with state-of-the-art AI. Supports multiple aspect ratios, styles, and high-resolution output for creative and commercial use.
PNG, JPEG, WebP, or GIF · 20 MiB maximum each
Example output — click Run to generate your own
Send your first request
OpenAI-compatible endpoint with unified authentication and usage tracking.
https://api.sandbase.ai/v1/runhidream-ai/hidream-o1/devInput Schema
5 parameters · 1 required · 4 optional
| Parameter | Type | Required | Description |
|---|---|---|---|
prompt | string | Required | Text prompt for image generation, editing, or subject-driven personalization. · Min length: 1 |
seed | integer | Optional | Random seed for reproducible generation. |
images | string[] | Optional | Optional reference images. Use none for text-to-image, one for image editing, or multiple for subject-driven personalization. · Default: [] |
aspect_ratio | string | Optional | The aspect ratio of the generated image. · Options: 21:9, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16 · Default: "16:9" 21:916:93:24:35:41:14:53:42:39:16 |
output_format | string | Optional | The format of the generated image. · Options: jpeg, png · Default: "png" jpegpng |
Output Schema
| Field | Type | Description |
|---|---|---|
id | string | Unique identifier for the generation task |
status | string | Task status: pending, running, completed, failed, timeout |
model | string | Model used for the generation |
outputs | array | Array of output items |
outputs[].url | string | URL of the generated artifact |
outputs[].content_type | string | MIME type (e.g. image/png, video/mp4) |
error | object | null | Error details if failed, null on success |
error.type | string | Machine-readable error type code |
error.message | string | Human-readable error description |
Async Workflow
This model uses asynchronous execution. Submit a request and poll for the result.
- Submit — POST to /v1/run, receive an
id - Poll — GET /v1/run/{id} until status is
completed,failed, ortimeout - Retrieve — Read
outputsfrom the completed response
Code Examples
Ready-to-run snippets
# Step 1: Submit
curl -X POST https://api.sandbase.ai/v1/run \
-H "Content-Type: application/json" \
-H "Authorization: Bearer YOUR_API_KEY" \
-d '{
"model": "hidream-ai/hidream-o1/dev",
"images": [
"https://static.sandbase.ai/examples/hidream-ai/hidream-o1/dev/input_reference_image_urls_0.png"
],
"prompt": "A cinematic product photo of a ceramic mug on a marble counter, soft window light, shallow depth of field.",
"aspect_ratio": "16:9",
"output_format": "png"
}'
# Step 2: Poll result (replace <id>)
curl https://api.sandbase.ai/v1/run/<id> \
-H "Authorization: Bearer YOUR_API_KEY"API README
HiDream O1 Dev
HiDream O1 Dev is a distilled image foundation model that unifies text-to-image generation, instruction-based editing, and subject-driven personalization. Its pixel-level unified architecture processes text, raw image content, and task conditions together for consistent in-context visual creation.
One endpoint covers three workflows: send no reference images to create from text, one image to apply an instructed edit, or multiple images to carry subjects into a new scene. This is the O1 Dev model, distinct from the earlier HiDream I1 family.
Highlights
- One model, three workflows. Moves between text-to-image generation, single-image instruction editing, and multi-reference personalization according to the supplied references.
- Pixel-level unified generation. Treats text, image pixels, and task conditions in a shared token space instead of separating generation and editing into unrelated systems.
- Subject-driven personalization. Uses multiple references to preserve a person, character, product, or visual identity while placing it in a newly described setting.
- Long-text rendering and layout control. Creates multi-region compositions with accurate multilingual text, supporting information-dense posters, editorial layouts, and designed graphics.
Pricing
| Unit | Price |
|---|---|
| Per request | $0.005000 |
When to Use
| Good fit | Consider alternatives |
|---|---|
| Creating an image from text with zero references | A specialized text-to-image route when references will never be used |
| Applying a natural-language change to one image | Pixel-perfect masks or manual retouching |
| Personalizing a scene from multiple subject references | A single-reference task with strict reference preservation |
| Keeping a character or product recognizable in a new setting | Video, animation, or 3D generation |
| Combining generation and editing behind one integration | Workflows requiring explicit layout boxes or skeleton controls not exposed here |
Prompt Guide
Match the prompt to the reference count. With no images, fully describe the scene. With one image, state exactly what should change and what must remain. With multiple images, identify the shared subject and describe the new setting, pose, camera, and identity details to preserve.
{
"prompt": "Use the referenced ceramic mug as the exact product identity. Place it on a pale stone café table beside an open sketchbook, soft morning window light from the left, eye-level product photography, shallow depth of field. Preserve the mug's shape, glaze color, handle, and printed emblem.",
"images": [
"https://example.com/references/mug-front.png",
"https://example.com/references/mug-side.png"
],
"aspect_ratio": "4:3",
"output_format": "png",
"seed": 42
}
Technical Specs
| Spec | Value |
|---|---|
| Model ID | hidream-ai/hidream-o1/dev |
| Input | Required text prompt; optional array of reference-image URLs |
| Reference workflow | 0 images: text-to-image; 1: editing; 2 or more: subject-driven personalization |
| Aspect ratios | 21:9, 16:9 (default), 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16 |
| Output format | PNG (default) or JPEG |
| Optional controls | Integer seed |
| Output | One downloadable image URL |
| Execution | Async |
Related
- HiDream O1 — Full O1 generation model
- HiDream O1 Dev Edit — Dedicated editing route for the Dev variant

