# HiDream O1 Image Generator and Editor

> One native model for generation, natural-language editing and subject-driven reference composition.

Canonical URL: https://allimageai.com/models/hidream-o1-image/

Website: All Image AI

Language: English

HiDream O1 Image is a unified image model: the same operation can start from text, revise one image or combine several subject references. It supports custom dimensions up to roughly 2K without switching to a separate editing model.

## Model details

- Developer: HiDream.ai
- Status on All Image AI: Active
- Last verified: 2026-08-25
- Maximum output: 2K
- Modes: Text to Image, Image to Image, Edit Image, Multi-reference
- Reference images: up to 3
- Resolutions: 1K, 2K
- Aspect ratios: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3

## Best for

- Natural-language image editing
- Subject-driven generation
- Multi-reference compositions
- High-resolution product concepts

## How to use

1. Generate from text or upload references.
2. Identify each reference by number.
3. State what must remain unchanged.
4. Use 2K only for the selected result.

## Settings explained

### Reference Images

Use one for editing or up to three for subject-driven composition.

### Guidance

The documented full-model default is 5; raise it cautiously for strict instructions.

## Prompt examples

### Subject composition

> Place the person from image 1 in the studio from image 2, holding the product from image 3; preserve identity and label details

### Local edit

> Keep the room and camera angle unchanged; replace only the chair with a low walnut lounge chair

## Limitations

- The interface caps references at three for predictable prompts
- The full 50-step model is slower than its distilled variants
- Dense copy still needs proofreading

## Pricing and credits

Credits for one image at default quality and mode. Resolution, quality, reference inputs, edit mode and output count can change the total. Check the Generate button before submitting.

| Resolution | Credits / image |
| --- | --- |
| 1K | 1 |
| 2K | 3 |

Failed generations cost 0 credits. Catalog availability does not guarantee runtime availability; the workspace reports whether the selected model can run.

[Credit plans](https://allimageai.com/pricing.md)

## Test results

Testing in progress. No verified benchmark scores have been published for this model.

## Commercial use and image handling

Commercial use is subject to the HiDream O1 model licence and the selected service terms.

Uploads are processed by the configured inference service. You can delete a generation from your history.

## Frequently asked questions

### Does editing call a different model?

No. This integration uses the unified HiDream O1 Image operation and passes references directly.

### Why only three references?

The upstream schema allows multiple images without publishing a numeric limit, so the interface uses a conservative cap.

## Release history and sources

May 8, 2026 — Open weights released

This date applies to the event shown. Later API access, weights and availability on All Image AI may have different dates.

- [github.com](https://github.com/HiDream-ai/HiDream-O1-Image)

[Event evidence on the timeline](https://allimageai.com/timeline/#event-hidream-o1-image)

## Other versions

- [HiDream I1 Full](https://allimageai.com/models/hidream-i1-full.md): Full quality · 2K
- [HiDream I1 Fast](https://allimageai.com/models/hidream-i1-fast.md): 16 steps · 2K