Midjourney
Kyeo's Midjourney page generates from text or up to four reference images, with three speeds, seven aspect ratios, and five V7, V6, or Niji versions. It is not a full mirror of Midjourney's product: Midjourney now defaults to V8.1, while Kyeo still defaults to V7 and exposes no Style Reference, Omni Reference, or video workflow.
Online use is currently unavailable
This page is for reviewing the model's capabilities, use cases, and public references. The model does not appear in the home workbench and cannot submit generation tasks.
At a glance
What is this model like?
A prompt of up to 2,000 characters uses `mj_txt2img` when no image is present and switches to `mj_img2img` with one to four uploads. Credits range from 3 for Relaxed to 16 for Turbo, while Fast costs 8 for text generation and 6 when an image is uploaded.
Key facts
Quickly assess whether this model matches your workflow.
Also known as
The same model may appear under different names across documentation and community discussions; this list keeps them easy to search and compare.
Common search intents
These questions and keywords combine public references with common creative use cases to help you choose faster.
Selection guide
Use these decision points when choosing a model.
Use the same brief across V7, V6.1, and Niji 7 when the task is to evaluate composition, mood, subject interpretation, or a more illustration-oriented treatment.
Compare Nano Banana Pro or Flux 2 Pro for prompt-led changes to an existing image, and Ideogram Character when one subject's recognizability across scenes is the core review.
Then hold V7 and 1:1 constant to compare speed; for a version test, hold the prompt and `speed` constant and switch only to V6.1 or Niji 7.
Popular comparisons
Common comparison questions help you find the practical differences quickly.
Model comparison
Compare the current model with alternatives at a glance.
Practical usage insights
Practical observations based on real creative workflows.
Midjourney's default moved on; Kyeo's version list did not
Kyeo credits and Midjourney GPU time are separate systems
The legacy workflow remains in code after its public docs disappeared
Capabilities
Separate text and reference-image paths
Three speeds, seven ratios, and five versions
Speed and input path determine credits
No role or weight controls for reference images
Use cases
Poster and cover concepts
Same-brief V7 and Niji comparison
Mood-board and sketch guidance
Prompt tips
Fix version, ratio, and speed first
Add reference images one at a time
Keep style references separate from scene instructions
Strengths
Limitations
Adjustable parameters
Quickly assess whether this model matches your workflow.
FAQ
Related models
Compare these similar candidates before deciding.
Nano Banana Pro
Nano Banana Pro supports text-to-image generation and reference-image editing on Kyeo AI, with 1K, 2K, or 4K output. Each 1K or 2K request costs 8 credits, while 4K costs 14. Resolution is an output-size tier, not a guarantee of quality, accuracy, or detail.
Flux 2 Pro
Flux 2 Pro on Kyeo supports text-to-image and image editing. A request without an upload uses `flux-2/pro-text-to-image`; attaching one to eight images switches to `flux-2/pro-image-to-image`. Both routes offer seven fixed aspect ratios and 1K/2K, currently priced at 5/7 credits.
Ideogram Character
Ideogram Character combines one character reference and a prompt to generate a new single image. It is built for testing whether a subject remains recognizable across scenes, poses, or compositions. Details can still drift, so consistency is a review target—not a guaranteed result.
Sources
Content is based on model documentation, feature references, and the settings available on Kyeo AI.
Midjourney's official version, speed, and reference documentation defines the available scope. Kyeo exposes only `mj_txt2img`, `mj_img2img`, `speed`, `aspectRatio`, `version`, and up to four images. Midjourney now defaults to V8.1, while Kyeo still defaults to V7 and offers no V8 or V8.1 option.
Historical Generate and Task Details schemas are not treated as proof of current availability. Technical identifiers: `llms.txt`.
Kyeo still sends `taskType`, prompt, speed, aspectRatio, version, and optional fileUrls to the upstream interface's `/mj/generate`, then polls a dedicated status workflow. Live success and the response shape require a representative sample.
Kyeo offers Text-to-Image and Image-to-Image with a 2,000-character prompt, up to four images, and seven ratios. It defaults to V7 and omits Midjourney's current V8.1 default; pricing follows `speed` and upload presence.