Midjourney Guides Hub

Updated: 2026-08-29. Default versions, parameter availability, subscription tiers, and access policies change. Treat midjourney.com and in-product docs/announcements as the source of truth for that day. Third-party entries are not Midjourney official products.

Introduction

Searches for “Midjourney tutorial,” “Midjourney guide,” or “Midjourney overview” often tangle three different things: entry (official Web, Discord, China-friendly third-party tools), model (V8.2 / V8.1 / Niji 7 / Video, and so on), and how you call it (web UI, bot commands, personalization settings). This page is the map for the whole Midjourney cluster—get the layers right, then jump to “what it is,” China access, V8.2, Niji, prompts, or styles without bouncing through nine articles blindly.

Important fact up front: Midjourney has no publicly available official developer API. These guides will not pretend there is an official HTTP API you can wire into a product. For product integration, evaluate compliance and Midjourney’s own open-access timeline separately—and never treat a third-party “compatible interface” as official capability.

What this guide solves

• A single table for what to learn, which guide to open first, and what to practice next
• Clear separation of entry, model, and personalization so UI button labels are never mistaken for “official API model IDs”
• A mental map of the Midjourney family: when V8.2, V8.1, Niji 7, or Video actually fit
• A five-step path to a reusable first render—not one vague adjective prompt
• Scenario-based routing for beginners, China access, flagship models, anime, prompts, and style libraries

Learning path at a glance

Goal	Read first	Practice next	Success signal
Understand what Midjourney is	Family explainer	Same brief on V8.2 / V8.1 / Niji	You can say when the flagship is unnecessary
Open and generate from China	China access guide	Bookmark a trusted entry; finish a first image	You can name third-party risks
Master the daily flagship	V8.2 full guide	Lock aspect ratio and personalization	Fewer redo rounds
Anime / illustration track	Niji 7 selection & practice	Character consistency and line-art constraints	Style drift stays controllable
Ship your first full render	Getting started	Checklist composition and detail	You can reproduce the same layout
Stabilize prompts	Prompt playbook	Build a personal template library	Same prompt behaves predictably
Capture reusable styles	Styles & scenes	Style cards by use case	Team shares one description set

In most cases, change the prompt, aspect ratio, and personalization before you chase a newer model version. Blindly chasing “whatever is newest by default” only makes failures harder to reproduce.

Three concepts to lock in first

Entry ≠ model ≠ “API”

Concept	What it is	Common mistake
Entry	Official Web, Discord bot, China-friendly third-party workbench	Thinking “opening a site” equals the latest official default model
Model	Sets aesthetic lean, detail, anime vs photo-real, video bounds	Judging version from marketing copy or button nicknames alone
How you call it	Web generate, Discord commands, personalization (Moodboard / Profile, etc.)	Treating a third-party “compatible API” as Midjourney’s official public API

• Official site: www.midjourney.com — preferred when you need default models, subscription, Create page, and announcements.
• Discord: historically important for commands and community; whether it is still open for your account and what command format looks like follows official and Discord guidance that day.
• China-friendly convenience (third-party): Midjourney V7 — third-party model page, not a Midjourney product; the listed model names are as-of that day (a platform may still show V7 while the official default is already V8.2).
• Mirror / text-to-image studio: Text-to-image studio — multi-model switcher; review terms and routing separately.

Third-party mirrors and aggregators may differ on accounts, log retention, and model routing. Before uploading sensitive assets, trademarks, or unpublished product art, read their terms. For formal selection and billing, return to midjourney.com.

Again: as of this guide’s writing, Midjourney has no official public API. If a platform claims “official API direct access,” verify the contracting party and disclosures yourself—do not write procurement specs that treat that as Midjourney capability.

Midjourney family map (2026)

Midjourney is a standalone image (and extended video) generation product—not a chat-bot side button. Public versions and modes shift with release cadence; the table below is for navigation only:

Product / mode name	One-line fit	Better for	Dig deeper
Midjourney V8.2	Common flagship default (confirm on site that day)	Photo-real, design-led, everyday finals where quality matters	V8.2 full guide
Midjourney V8.1	Same-gen prior tier / comparison baseline	Aligning with existing workflows; A/B against the neighbor version	Family explainer
Niji 7	Anime / comic-illustration lean	Characters, manga feel, Japanese-illustration aesthetics	Niji 7
Video (related capabilities)	Still frames into short motion / video drafts (per site that day)	Motion previews or short-shot concepts	Official feature pages that day

Whether a model appears in your account region, renames, or changes default is always per the official site that day. A “Midjourney” label on a third-party page does not guarantee a particular official weight on the backend.

Output quality follows constraints, not only the version number

Under the same model, spelling out subject, lens, light, aspect ratio, and bans usually cuts rework more than blindly jumping to a newer version. Brand marks, lookalike faces, medical ads, and imagery of minors also need compliance and human review—smooth generation is not a green light to publish.

Treat “one generate” as a managed loop: short prompt to lock composition → add style and light → only then add on-image text or complex props. Change one variable per round so you know whether the prompt or the model tier failed. For teams, write acceptance criteria (safe margins, readable title, bans)—“make it prettier” is not actionable.

Personalization (aesthetic preference, Moodboard, custom profiles) can sharply change the result distribution of the same prompt. Before you switch devices or accounts, record the current Profile / settings snapshot—otherwise “it worked yesterday, everything drifts today” is almost impossible to debug.

Treat personalization as versionable project config: exploration can switch freely; delivery locks a verified combo and notes “model version + personalization on/off + aspect” in captions or tickets. Three months later you can still explain why a frame looked that way. Viral “universal prompts” usually assume someone’s personalization stack; if your account does not share it, paste-and-pray fails—not because you cannot write prompts, but because the config layer never matched.

Subscriptions, credits, and “speed tiers” (no hard-coded numbers)

Midjourney usage usually ties to subscription tier, generate mode (more economical or faster queues—names follow the UI), and whether you upscale/redraw. These guides do not freeze monthly fees or per-image credits. Build two habits: drafts stay cheap; finals get the upscale—and external quotes never hard-code “one image = N credits”, because rules change. Billing disputes follow the official subscription page and email receipts.

Five steps to a strong first render

1. State use and aspect: poster, cover, icon, illustration; landscape / portrait / square; target channel (web hero, short-video cover, and so on).
2. Name subject and scene: who or what sits at center; environment, time, weather, props—avoid adjective piles alone.
3. Specify style and camera: photo-real, concept art, product studio, flat illustration; framing (close / mid / wide) and light language (side light, backlight, depth of field).
4. List bans and text rules: no watermarks, no garbled English, no extra limbs; if on-image text is required, quote the exact string and ask for clear, readable glyphs—important titles are often safer in post.
5. Human-check, then iterate: match use case for composition, readability, brand consistency; change one variable (light or framing) so failures are diagnosable. Whether Upscale / Remix / Vary exist and what they are named follows the UI that day.

Copy-ready first-image template:

Use: [web hero / cover / product concept]
Aspect: [16:9 / 1:1 / 9:16]
Subject: [clear person or object, including material and color]
Scene: [place, time, mood]
Style: [photo-real / concept art / other concrete style]
Camera: [framing + lighting]
On-image text (if any): "exact copy"; must be sharp and typo-free
Ban: watermarks, extra fingers, warped logos, cluttered background text
Output: one primary image plan; if anything is unclear, ask before guessing brand rules.

Privacy-safe comparison prompt (quick entry check):

Generate a photo-real still life: a coffee cup and a closed notebook on a desk.
Natural window light, shallow depth of field, no watermark, no close-up faces.
Do not add any on-image text.

Run the same prompt on the official Create page / Discord (if available to you) / any China-friendly third-party entry you have evaluated. Log speed, quality, and whether the model invents text—then pick a long-term path.

First-render failure cheat sheet:

Symptom	Likely cause	Try first
Subject OK, crop kills it	Aspect missing or buried at the end	Put aspect first; reserve title safe zone
Beautiful but disobedient	Personalization too strong or stylize too high	Turn off / lower personalization, compare again
Fingers / edges collapse	Complex hands or heavy occlusion	Simplify pose, or fix locally in post
Invented / garbled text	On-image text not banned	Explicit “no text” or type in post
Third-party far from official	Different version routing	Log both model names; deliver via official

Pin this table in the team wiki—it cuts time spent “switching entries and hitting the same wall.”

Topic map by scenario

What you want now	Open this	Have ready
First concepts	What is Midjourney	Three real image briefs
Official site won’t open / need a fast try	China access	A privacy-safe test prompt (no trademarks)
Systematically master flagship V8.2	V8.2 guide	Fixed aspect + personalization notes
Anime / character illustration	Niji 7	Short character identity paragraph
Hand-held first image	Getting started	This page’s five-step template
Prompts keep collapsing	Prompt playbook	Personal negative list
Reusable style library	Styles & scenes	Brand colors + licensed refs

Access & entries

• Official Midjourney: www.midjourney.com
• China-friendly convenience (third-party · model page example): Midjourney V7
• Mirror / text-to-image studio: Text-to-image studio

Note: third-party ≠ official; whtai model lists are as-of that day and may still show V7 while the official default is V8.2. For important delivery, cross-check one privacy-safe prompt on the official path. Before uploading enterprise assets, portraits, or unpublished designs, confirm terms and your org’s data policy.

Learning kickoff checklist

• You can separate entry, model, and personalization—and will not treat third-party button names as an official public API
• You bookmarked at least one trusted entry (official, or a China-friendly entry whose terms you reviewed)
• You finished one “five-step render” with a privacy-safe fixed prompt and human review
• You know whether the next job is selection, China access, V8.2, Niji, prompts, or style capture—and opened that guide
• You treat key frames as requiring human verification; first generate is never the final
• Before formal selection or purchase, you checked Midjourney’s official pages and announcements—not only third-party labels
• You know: no official public API; “compatible interfaces” are not Midjourney commitments

Quick access

• China-friendly entry (third-party): Midjourney V7
• Mirror studio: Text-to-image studio
• Official site: www.midjourney.com

FAQ

Is Midjourney free?

Subscription tiers, trials, and credit rules change. Follow midjourney.com subscription and billing copy that day—do not rely on expired screenshots or second-hand quotes. This guide does not hard-code prices.

V8.2 or Niji 7?

For general photo-real, product concepts, and design-led visuals, start with V8.2 (or the official default flagship that day). For clear anime / comic-illustration delivery, open Niji 7. Details live in the family explainer and each dedicated guide.

Are China third-party entries the same as official?

No. Third parties own reach and packaging; accounts, logs, model routing, and support channels can differ. When you need product capability and default version truth, return to midjourney.com.

Does Midjourney have an official public API?

There is no publicly available official developer API (per Midjourney’s statements that day). Do not assume an official REST surface for automation; third-party wrappers need separate legal and technical review.

Can I use generated images commercially?

It depends on the product terms you accepted, regional rules, asset rights, and image content (trademarks, likeness, brand assets). Treat the model as a draft accelerator; finish rights and compliance review before commercial use. These guides are not legal advice.

Which guide should I read first?

Complete beginners: this page → What is Midjourney → China access or Getting started. Stable entry already, just need quality: go straight to the V8.2 guide or Prompt playbook.

Official resources

• midjourney.com
• In-product docs, announcements, and subscription notes (whatever is visible after login that day)

Further reading

• What is Midjourney? Family map
• Midjourney China access guide
• Midjourney V8.2 full guide
• Midjourney Niji 7 selection & practice
• Midjourney getting started: first image
• Midjourney prompt playbook
• Midjourney styles & scenes

Summary

The right order is: pick a trusted entry → choose V8.2 / V8.1 / Niji by task → control quality with structured prompts, personalization, and human review. Use this page as a table of contents: finish one real render job, then open the matching deep guide—faster than reading everything at once. Entry is not model; third-party is not official; there is no official public API; aging defaults and parameters always follow Midjourney’s site that day.