Midjourney

Image GenerationReference GuidedImage Editing

Midjourney through APIXO is a multi-version image-generation route centered on the current V8.1 model. It supports prompt-based creation, image guidance, dedicated image editing, compatible character, style, and object references, plus follow-up Vary and Upscale actions for production-oriented visual exploration.

APIXO Versions

V8.1 / V7 / V6.1 / V6 / Niji

Price

From $0.10 / UseDraft to Turbo

V8.1 HD

Output

4 Variations

V8.1 Released

Apr 30, 2026

Loading workspace...

Capabilities

Creative Direction Beyond a Single Text Prompt

V8.1 Prompt Interpretation

Midjourney V8.1 improves prompt reading and retention of small visual details compared with earlier versions. This benefits scenes containing specific materials, lighting relationships, accessories, background elements, and carefully described compositions.

Native 2K HD

APIXO exposes V8.1’s HD option for text-to-image and image-to-image generation. It creates a higher-resolution 2K result directly rather than requiring the separate Upscale action used for selected grid outputs.

Faster V8.1 Rendering

Midjourney reports that standard V8.1 jobs render approximately four to five times faster than earlier versions. Actual APIXO completion time still varies with the selected mode, output settings, and provider conditions.

Style References

Compatible versions can use Style References to transfer visual qualities such as color, medium, texture, lighting, and overall treatment. Style weight adjusts influence without treating the reference as a source-object copy.

Version-Specific References

APIXO exposes character, style, and object reference fields on compatible versions. Character Reference behavior belongs to earlier Midjourney versions, while V7 uses Omni-style subject guidance. V8.1 does not natively support either subject-reference system, so reference availability must be matched to the selected version.

Aesthetic Range Controls

Adjust stylization, variety, weirdness, and supported experimental behavior to move between literal prompt interpretation and more exploratory results. Negative prompting can reduce unwanted concepts in text-to-image and image-to-image workflows.

Workflows

Generate New Directions or Refine an Existing Result

Prompt and Reference Generation

Use text-to-image for prompt-led creation or image-to-image for general visual guidance. Compatible versions can add image prompts, Style References, character references, or one object reference, while aspect ratio and aesthetic controls shape the new four-image result.

Create

Editing and Result Actions

Image-edit accepts one source image and a prompt for dedicated modification. Vary and Upscale instead require a successful prior APIXO task and a selected image index, keeping follow-up work tied to one result from the original four-image grid.

Refine

Specifications

Midjourney Options Available on APIXO

Version compatibility determines which references and controls can be combined.

V8.1 / 7 / 6.1 / 6 / Niji 7 / Niji 6

Versions

Text / Image / Edit

Creation Modes

Vary / Upscale

Result Actions

11 Presets

Aspect Ratios

Draft / Fast / Turbo

Speed Options

V8.1 HD

Use Cases

Production Workflows for Midjourney on APIXO

Campaign Development

Visual Direction Boards

Explore four interpretations of a campaign brief in each initial generation. Style References and adjustable stylization help creative teams compare photographic, illustrative, editorial, and surreal directions before choosing a visual language for further production.

Brand Exploration

Cohesive Content Series

Reuse an approved Style Reference across related concepts to guide palette, texture, lighting, and medium. This supports visually connected campaign assets while leaving room for different subjects and compositions across the series.

Product Concepts

Art-Directed Product Scenes

Combine a product or object reference with a descriptive environment, composition, and lighting brief to explore advertising concepts. Final geometry, packaging text, logos, and regulated product details still require comparison with approved source assets.

Illustration

Editorial and Narrative Artwork

Use Midjourney or a compatible Niji version to explore covers, key art, character concepts, environments, and editorial illustrations. Version selection keeps general visual generation separate from Niji’s anime- and illustration-oriented model family.

Notes & FAQ

Midjourney Version and Workflow Guidance

Important Notes

APIXO’s route exposes multiple Midjourney versions; V8.1 is the current upstream default, not the only selectable model.

V8.1 does not natively support Character Reference or Omni Reference. Character, object, and style-reference availability depends on the selected Midjourney version.

Image-edit requires exactly one source image and should not be described as mask-based inpainting or outpainting.

APIXO reports typical generation times of 30–60 seconds, but completion time can vary with mode, output, and provider conditions.

References guide a generation rather than guaranteeing exact identity, typography, object geometry, or deterministic reproduction.

APIXO does not publicly expose transparent backgrounds, vector files, layers, pan, zoom-out, or standalone mask controls on this route.

Frequently Asked Questions

The route is multi-version rather than fixed to one checkpoint. APIXO exposes V8.1, V7, V6.1, V6, Niji 7, and Niji 6. V8.1 was released on April 30, 2026 and became Midjourney’s upstream default on June 10, 2026.

A new text-to-image, image-to-image, or image-edit use returns four variations. Vary and Upscale operate on one selected image from a prior successful task and are billed as separate follow-up uses.

Text and image generation cost $0.10 in Draft, $0.20 in Fast, $0.30 for V8.1 HD Fast, or $0.40 in Turbo. Image-edit costs $0.30 in Fast or $0.60 in Turbo; Vary and Upscale each cost $0.30 per use.

No. APIXO exposes Draft, Fast, and Turbo as route-level options, but not every speed is compatible with every version. Midjourney currently does not support Turbo generation with V8.1. Select only combinations supported by the chosen version.

Short Latin-alphabet text can be prompted by placing the requested wording in double quotation marks, but exact spelling and layout are not guaranteed. Logos, packaging, identity-sensitive subjects, hands, and commercial claims should receive human review before publication.