Overview
HiDream O1
HiDream O1 is an open-weight image model built around a pixel-level unified transformer. Its published model family supports generation, editing, and subject-driven workflows up to 2048 × 2048.
This MachGen catalog entry currently exposes text-to-image only. Editing and subject-driven conditioning are not available in this generation surface.
Model guideWhy choose HiDream O1
Key strengths
- High-detail text-to-image generation across photoreal, illustrative, fantasy, and design-led styles.
- Large fixed output sizes support dense texture and more complex compositions.
- Seed control makes it easier to refine a chosen visual direction.
Good fit for
- Fantasy environments, character concepts, editorial illustration, and polished stills.
- Moodboards and visual development that need rich material or lighting detail.
- Repeatable composition studies from a text brief.
Supported on MachGenInputs and output settings
This table reflects the model routes and controls currently exposed by MachGen.
| Mode | Required input | Available settings |
|---|
| Text to Image | Text prompt | Output sizes: 1024x768, 1024x1024, 1024x1808, 2048x2048, 2048x1536 Seed control |
WorkflowUsing HiDream O1
- Choose a supported generation mode in Playground.
- Add the required prompt and source or reference assets.
- Select output settings from the controls shown for that mode.
- Generate, then inspect the returned asset before reusing the settings through the API.
Practical guidance
- Describe the subject, action or composition, environment, and visual direction in a clear order.
Capabilities
What you can make
Text to image
Write a full scene brief - subject, environment, materials, and lighting - to get high-detail stills from fantasy worlds to everyday photography.
Style range
Works across illustrative and photoreal aesthetics; be explicit about medium (oil, anime, product photography) when the look must stay locked.
Prompting
Prompting notes
- Lead with the subject, then environment, then lighting and camera.
- Call out texture and material early if they define the shot.
- Keep conflicting styles out of the same prompt unless you want a hybrid look.
Best resultsHow to get better HiDream O1 results
- Start with subject and composition, then add material, lighting, environment, and style.
- Use camera and lens language only when a photographic result is intended.
- Avoid combining several unrelated art styles; select one medium and describe its defining traits.
- Keep the seed while refining detail, then change it when the underlying composition is not working.
LimitsLimits and availability
- MachGen admits only the modes and settings listed above for this catalog model.
- Source assets must finish uploading before a request can be submitted.
- Generated results can vary between requests, including when the same prompt and settings are reused.
- HiDream O1 runs on MachGen-managed inference infrastructure rather than being forwarded to a provider API.
Learn moreLearn more
For model background, technical details, and architecture, see the official HiDream O1 repository.
Try this model in the MachGen Playground.
Official documentation may describe capabilities or parameters that are not currently exposed by MachGen. Use the Inputs and output settings section above as the source of truth for this page.
PricingHiDream O1 pricing
See Pricing for currently published MachGen rates. When available, the Playground estimate reflects the selected task and output settings.
On MachGenHiDream O1 on MachGen
Deployment: MachGen Hosted. HiDream O1 runs on MachGen-managed inference infrastructure rather than being forwarded to a provider API.
The Playground and API sections on this page list the inputs and output settings exposed for this model.