Midjourney V8.1 Style Transfer API by MIDJOURNEY
Midjourney retexture changes the artistic style of an input image while preserving its composition, returning four restyled results.
1. Introduction
Midjourney V8.1 Style Transfer (midjourney/v8.1/style-transfer) regenerates an input image in a new style described by a text prompt while preserving the original's structure and composition. Built on Midjourney's Retexture capability, it "re-skins" a scene — re-estimating its shapes and re-applying materials, surface qualities, and lighting — so the layout stays intact while the look changes entirely. Each task returns four restyled variations.
It is part of the Midjourney V8.1 family exposed through this API:
midjourney/v8.1/text-to-image— generate from a text promptmidjourney/v8.1/image-to-image— generate guided by one input imagemidjourney/v8.1/blend— fuse 2–5 imagesmidjourney/v8.1/style-transfer— restyle an image, preserving composition (this model)midjourney/v8.1/remove-background— isolate the subject on transparencymidjourney/v8.1/image-to-video— animate an image into a short clip
Midjourney V8.1 is built by Midjourney, Inc., an independent, self-funded San Francisco research lab founded in August 2021 by David Holz.
2. Key Features & Innovations
- Composition-preserving restyle: The whole image is regenerated in a new style while the structure, layout, and composition of the original are retained — like applying a fresh skin to a fixed scaffold.
- Text-driven styling: The target look is specified in plain language (e.g. "cyberpunk neon", "marble sculpture", "watercolor"), making it easy to A/B many styles over the same composition.
- Distinct from style reference: Midjourney's Style Reference (
--sref) copies only the style of a reference image and generates new content; Style Transfer instead keeps your image's content/geometry and changes its style. It also differs from image-to-image, which treats the input as loose inspiration rather than a structural template. - Built on Retexture (Midjourney v6.1+): This endpoint runs on V8.1.
- Four results per request: Each task returns a 4-image grid of restyled candidates.
3. Parameters & Usage
| Parameter | Description |
|---|---|
image (required) | Input image to restyle. Provide a publicly reachable HTTPS URL or upload. Composition and structure are preserved from this image. |
prompt (required) | Text describing the target style, material, or medium to apply (e.g. "ukiyo-e woodblock print", "brushed metal", "low-poly 3D render"). |
Tips: Describe the desired material/medium and overall aesthetic rather than the scene contents — the contents come from the input image. Because geometry is preserved, Style Transfer is well suited to producing consistent style variants of a fixed layout.
4. Model Architecture & Technical Details
Midjourney V8.1 is a complete from-scratch rewrite of the company's image model, built on a GPU-native PyTorch stack (migrated from the earlier TPU infrastructure). The underlying generative approach is understood to be latent diffusion; Midjourney has not published a technical paper or model card, so the backbone, parameter count, and training data remain undisclosed. The defining methodology is a human-preference (RLHF-style) aesthetic tuning loop combined with per-user personalization. V8.1 was released on midjourney.com on April 30, 2026 and became the default Midjourney model on June 10, 2026.
Retexture is a Midjourney editing capability introduced in v6.1 and is largely independent of the base model version. The training dataset has never been disclosed and is the subject of active, unresolved copyright litigation — Disney Enterprises, Inc. v. Midjourney, Inc. (No. 2:25-cv-05275, C.D. Cal.), filed June 11, 2025 by a coalition of major studios including Disney, Marvel, Lucasfilm, Twentieth Century, Universal, and DreamWorks. Those infringement claims are allegations in pending litigation and have not been adjudicated.
5. Intended Use & Applications
- Material & style studies: Generate many material/finish variants of a single design or product mockup.
- Reskinning concepts: Re-render a scene or character in alternative art styles while keeping its blocking.
- Consistent layout, varied look: Produce style A/Bs of a fixed composition for design reviews.
- Restyling photos & renders: Convert a photo or 3D render into an illustrated or stylized treatment.
For loose, inspiration-driven restyling that may change composition, use midjourney/v8.1/image-to-image; to apply only the style of a reference image to brand-new content, use the sref style-reference parameter on text-to-image / image-to-image.
