About Midjourney
Midjourney is an image generator driven by written prompts, with a companion video model that turns a still image into a short clip. It runs on Discord and in a web app at midjourney.com.
Prompts are shaped with parameters. Version 8 supports chaos, weird, exp, raw, hd and quality, and a stylize value of 1000 is the recommended setting there for aesthetic results. HD output is chosen either in the settings panel or by adding an hd flag to a prompt, and with that flag the model renders natively at 2K resolution. Text inside an image comes out more accurately when the words you want are wrapped in quotation marks in the prompt. niji 7 is Midjourney's anime image model and works with the same personalization and moodboard features as the main line.
Personalization is what separates one account's output from another's. A profile is built by clicking through a scrolling list of images rather than by comparing pairs, and it activates once 40 images have been rated. It stabilises at around 200 ratings and keeps improving up to roughly 2,000. You can hold several named profiles, set one or more of them as a default, and filter your image history by the profile that produced a given picture. Moodboards work alongside profiles: you upload a collection of images that the model then remixes, and a more varied collection produces more complex blending. Version 8 stays backwards compatible with profiles, moodboards and style references made under version 7, and version 8.2 draws its profiles from a wider pool of images than the previous generation.
Read moreShow less
Video starts from a still image. An Animate button offers an automatic mode or a manual mode driven by a motion prompt, along with a low motion setting suited to ambient scenes and a high motion setting for moving subjects and moving cameras. A single job returns four clips of about five seconds each, and a clip can be extended by roughly four seconds at a time, up to four times. Video is the expensive part of the service. A video job costs about eight times an image job, which works out at roughly the cost of one image per second of footage.
The web app is built for people producing images in volume. The aspect ratio slider includes 6:11, 4:5, 5:4 and 21:9, and bulk actions such as download, like, trash, publish and folder moves handle up to 2,000 items at once.

Key features
- Generates images from text prompts, with niji 7 as a dedicated anime model.
- Renders natively at 2K resolution when the hd flag is set from the prompt or the settings panel.
- Accepts chaos, weird, exp, raw, hd and quality parameters on a prompt.
- Renders requested wording more accurately when it is placed inside quotation marks in the prompt.
- Builds a personalization profile from rated images, active at 40 ratings and stable near 200.
- Holds several named personalization profiles, sets defaults and filters images by the profile used.
- Remixes user uploaded moodboards, blending more complexly as the images get more varied.
- Animates a still image with an Animate button, returning four clips of about 5 seconds each.
- Drives video automatically or from a written motion prompt, with low and high motion settings.
- Extends a generated video by roughly 4 seconds at a time, up to four times.
- Offers an aspect ratio slider covering 6:11, 4:5, 5:4 and 21:9 on the web app.
- Handles bulk download, like, trash, publish and folder moves for up to 2,000 items at once.
- Runs on Discord as well as on midjourney.com.
What people use it for
Key facts
| Category | AI |
|---|---|
| Pricing | Paid |
| Free tier | No |
| Upvotes | 7 |
| Listed | 22 August 2026 |
Is this your product?
Midjourney alternatives
- Open source
- Freemium
- Paid
- Freemium
More in AI
Run language models on your own machine with one command.
Speech synthesis and voice cloning that survives being listened to closely.
The model hub. Weights, datasets and demos for most of open-source ML.
OpenAI's assistant, with browsing, images and code execution built in.
An answer engine that cites its sources instead of listing ten blue links.
Video generation and editing built for people who already edit video.
Anthropic's assistant, strongest on long documents and careful reasoning.