Midjourney

Midjourney is an image generation service built by an independent research lab. It is closed, cannot be self-hosted, and is available only by subscription. In the early days you could only use it by typing commands in Discord; it now has a web app, and most of the work happens in the browser.

Judged purely on how the images look, it is still the most reliable: lighting, composition and material all land well from an ordinary prompt. For concept art, posters and mood references it saves the most time in sifting through results.

The cost is control. Precise instructions — this object in the lower left, that figure facing right — often produce a more beautiful image that is not the one you asked for. For work that must match a design exactly, the Stable Diffusion route with stackable conditioning fits better.

Key features

• Text-to-image: Each prompt returns several candidates; you vary or upscale the one you like and converge step by step.
• Style reference: Point it at a reference image and new results borrow its palette and look — useful for keeping a series consistent.
• Region editing and outpainting: Regenerate a selected area or extend the canvas outward, so small fixes do not mean starting over.
• Personalisation: Rate images and the model learns your taste, nudging later results toward it.
• Image-to-video: Turn a generated still into a short animated clip for mood pieces.

How to use

1. Go to midjourney.com, sign in with Discord or Google, and pick a subscription plan.
2. Type an English prompt describing subject, setting, style and lighting, then submit.
3. Choose a candidate and vary, upscale or edit it; add a reference image to steer the style.
4. Download the result. If the work is confidential, check whether your plan supports stealth mode.

Best for

• Concept design and mood boards: lots of high-quality directions fast, for aligning with a team or client.
• Posters and cover art: where atmosphere matters more than precise layout, its hit rate is the highest.
• Illustration and style exploration: style references let you try many looks far faster than hunting for references by hand.
• Personal projects: the lowest-friction way to get good-looking images without setting anything up.

Strengths and limitations

Strengths

• A high aesthetic floor — even plain prompts get solid composition and lighting.
• The web app is easy to pick up, with no GPU and no parameter tuning required.
• Style references and personalisation make a consistent series easier.

Limitations

• Weak control: exact placement, poses and typography are hard to get right in one go.
• Subscription only, with no lasting free tier; images on lower plans are public by default.
• Copyright questions around training data are unresolved; read the terms and weigh the risk before commercial use.
• Not reachable from mainland China directly.