Product profile
Editorially ResearchedMidjourney
High-quality AI image generation for concept art, brand exploration, and creative production.

Midjourney
Fit
Quick verdict
Best for
Creatives who want strong aesthetic image generation and iterative visual exploration.
Not ideal for
Teams that need fully automated brand-safe production pipelines with deep Figma-native editing, or buyers who require free unlimited commercial generation without reviewing license terms.
Primary strength
Distinctive, high-aesthetic image generation quality that many creatives prefer for ideation and visual storytelling.
Primary limitation
Workflow is generation-centric rather than a full professional design suite; brand consistency and legal/commercial terms still need process discipline.
Profile
Overview
Midjourney is a leading generative image platform known for distinctive visual quality, used by designers, marketers, and creators to produce concept art, campaign imagery, and creative explorations.
What sets it apartA generation-first creative tool prized for aesthetic image quality and iterative visual exploration.
Quick Verdict: TL;DR
Midjourney is still the aesthetic king of AI image generators, but it’s no longer the only serious option. V8.2, released July 24, 2026, leans harder into bold, opinionated visuals and personal taste than any version before it. Pricing didn’t change: $10 to $120 a month, with a 20% discount if you pay annually. The big 2026 plot twist isn’t a feature. It’s a business one. Midjourney just bought Co-Star, the astrology app, and hired its founder Banu Guler as Chief Design Officer. If you want gorgeous, cinematic imagery and don’t mind a manual workflow, Midjourney V8.2 is still worth paying for. If you want cheap text-in-image, easy prompting, or a public API, look elsewhere.
What Midjourney Is (and Isn’t) in 2026
Midjourney is a closed, hosted text-to-image model from Midjourney, Inc., a self-funded San Francisco lab of roughly 60 people. You feed it a text prompt, and it returns a 2x2 grid of images you can refine, upscale, remix, or reroll. That’s the whole product, and it does that one thing very well.
It is not a full design suite. There’s no Figma integration. There’s no layout engine. There’s no native way to lock down hex-accurate brand colors. There’s also no public API. Every generation goes through Midjourney’s web app at midjourney.com or its Discord bot. If you’re a developer looking to bolt Midjourney into a product pipeline, you cannot. Third-party “Midjourney APIs” are unofficial reverse-engineered Discord bots that violate the Terms of Service and risk account bans.
It’s also not great at text. V8.2 improves aesthetics and personalization but does nothing to fix Midjourney’s long-running weakness at rendering readable words inside an image. For signage, packaging mockups, or anything where the copy actually matters, ChatGPT’s image tools or Google’s Nano Banana Pro still beat it.
What it is is a fast, opinionated visual direction engine. Drop a few adjectives in, get four cinematic options in seconds, refine the one you like. That workflow hasn’t been beaten for moodboards, concept art, and editorial illustration.
Who Should Use Midjourney in 2026
Midjourney is a strong fit if you are:
- A creative director or art director exploring campaign visual directions before commissioning a shoot or illustration.
- A designer building moodboards and concept decks fast.
- A founder or marketer who needs hero imagery, blog headers, or social posts with a distinctive style.
- An author, game designer, or worldbuilder prototyping characters, environments, and props.
- A studio or agency running iterative visual exploration for clients and willing to spend on Fast GPU time.
It’s a poor fit if you are:
- A developer who needs programmatic image generation through an API. (It doesn’t exist.)
- A brand that needs every image to be on-system in Figma or Adobe Express with locked brand tokens.
- A team that needs a free tier. Midjourney killed free trials in March 2023 and hasn’t brought them back.
- Anyone producing final-deliverable print files. Midjourney outputs are starting points, not press-ready art.
Latest Models & 2026 Updates
The 2026 release cadence has been steady. Here’s what shipped, in order.
- March 17, 2026 — V8 alpha. First alpha of the V8 line drops.
- April 14, 2026 — V8.1 alpha. The version that becomes the public default.
- April 30, 2026 — V8.1 quality pass. Sharper output, especially on SREFs and moodboards; available on Discord and midjourney.com.
- June 11, 2026 — V8.1 becomes default. Replaces V7 across the platform. Native 2K HD mode, smarter prompt adherence, and much better in-image text rendering than V7.
- June 16, 2026 — Draft mode for V8.1. Each prompt generates 24 low-resolution images at half the GPU cost. Use Vary on the keepers to render at full quality.
- June 17, 2026 — Hardware announcement. First “secret hardware project” livestream. Details were teased but not yet shipped as a consumer product.
- June 25, 2026 —
--sref randomin draft mode. One prompt, 24 different style references applied. The fastest way to explore the style space. - June 25, 2026 —
--previewflag. A peek at the in-progress V8.2 model. - July 23, 2026 — Co-Star acquisition. Midjourney’s first-ever acquisition; Banu Guler joins as Chief Design Officer.
- July 24, 2026 — V8.2 goes live. The new default model. Bolder aesthetics, sharper detail, dramatically fewer random bad images, smarter personalization.
Midjourney’s own description of V8.2 says images should now feel “more creative, bold, sophisticated, edgy and fresh.” Random low-quality outputs, the long-running V8.1 complaint, are “dramatically reduced.” Personalization profiles, which use your image ratings to bias outputs toward your taste, now pull from a much larger and improved image pool, and they understand your personal taste better, especially if you’ve rated thousands of images.
V8.2 does not change pricing. It does not add a public API. It does not improve text-in-image rendering. It is a polish update, not an architecture rewrite. Most third-party testers find V8.2 a noticeable but incremental step up from V8.1, with the largest gains on complex multi-subject scenes where V8.1 used to fail.
Pricing Breakdown July 2026
Midjourney runs four subscription tiers. Annual billing cuts each by 20%, so the real entry price is $8 per month.
| Plan | Monthly | Annual (per month) | Fast GPU Time | Relax Mode | Stealth Mode | Concurrent Prompts (image) | Best For |
|---|---|---|---|---|---|---|---|
| Basic | $10 | $8 | ~3.3 hr/mo | None | No | 3 Fast | Casual users, learning prompts |
| Standard | $30 | $24 | 15 hr/mo | Unlimited images | No | 3 Fast or Relax | Solo creators, freelancers |
| Pro | $60 | $48 | 30 hr/mo | Unlimited images + SD video | Yes | 12 Fast or 3 Relax | Small studios, agencies, commercial work |
| Mega | $120 | $96 | 60 hr/mo | Unlimited images + SD video | Yes | 12 Fast or 3 Relax | High-volume teams, brand production |
A few important details:
- GPU time is GPU time, not wall time. One standard SD prompt (4 images) costs roughly 0.8 minutes of Fast GPU. HD quadruples the image resolution and roughly doubles the cost. Video is far more expensive: a batch of 4 SD videos is about 8 minutes; HD video is about 26 minutes.
- Unused Fast time does not roll over. Whatever you don’t use resets at renewal.
- Extra Fast time costs $4 per hour if you run out mid-cycle.
- Turbo mode is not currently supported in V8.1 or V8.2. Turbo doubles the Fast cost per image and was a feature on older models.
- The $1M revenue rule. Under the May 27, 2026 Terms of Service, any company (or employee of a company) earning more than $1,000,000 USD in gross revenue per year must be subscribed to Pro or Mega to own the assets they generate.
- Stealth Mode keeps your images and prompts out of the public Midjourney feed and out of remix rights for other users. It’s Pro and Mega only. For agencies, brands, and unannounced creative work, this is the must-have feature.
- Refunds are only available if you’ve used fewer than 20 GPU minutes across the lifetime of your account.
Midjourney does still work through Discord. Every plan lets you generate on both the website and Discord. The 1,000-image requirement to use the web app, a leftover from the Discord-only era, has been removed.
Midjourney vs Alternatives in 2026
Here’s how Midjourney V8.2 stacks up against the other major AI image tools that creatives are using in July 2026.
| Feature | Midjourney V8.2 | DALL-E / GPT Image (OpenAI) | Stable Diffusion 3.5 / Flux | Adobe Firefly |
|---|---|---|---|---|
| Aesthetic quality | Best in class for cinematic, stylized work | Very good, more “literal” | Open-source, varies by checkpoint | Strong on commercial photo, weaker on stylized art |
| Text in image | Weak (unchanged from V8.1) | Best in class for readable text | Improving but inconsistent | Good |
| Public API | None | Yes (OpenAI API) | Yes (self-host or Replicate) | Yes (Adobe API + Firefly in Creative Cloud) |
| Pricing model | Subscription tiers, $10–$120/mo | Pay-per-image, ~$0.04–$0.25 | Self-host free (GPU costs) or hosted per-call | Bundled with Adobe plans + credit packs |
| Style consistency across prompts | Strong via Style Reference and Personalization | Improving with custom instructions | Varies; LoRAs help | Brand-locked via Adobe Brandkit |
| Closed vs open | Closed | Closed | Open weights | Closed |
| Best for | Cinematic concept art, moodboards, brand exploration | Text-in-image, chat-integrated iteration, product mockups | Custom workflows, on-prem control | Enterprise brand-safe production inside Adobe stack |
| Notable 2026 weakness | No API, manual workflow | Less artistic polish than MJ | Higher setup cost | Less visually striking than MJ |
If your priority is the most beautiful raw image and you don’t mind a manual Discord-or-web workflow, Midjourney still wins. If you need text in the image, programmatic generation, or full Adobe-stack integration, look at the alternatives.
Real-World Use Cases
Where Midjourney genuinely earns its subscription in 2026:
- Editorial and campaign moodboards. Type a vibe, get four cinematic options in 4 seconds, present three to the client, refine the chosen direction. Midjourney’s HD mode renders at native 2K, which is more than enough for client decks.
- Character and environment concept art. The Personalization profile, especially with thousands of ratings behind it, gives you remarkably consistent “taste” across batches. Useful for indie authors, TTRPG publishers, and game studios.
- Brand visual exploration. Style References (
--sref) let you pin a look and apply it across many prompts. Stealth Mode on Pro/Mega keeps drafts private. - Architectural and interior mood imagery. Architects have used Midjourney since launch as a faster moodboard replacement for stock photo searches.
- Thumbnail and cover ideation. Bloggers and YouTubers use Midjourney to nail a thumbnail direction before commissioning a final illustration.
- Product and packaging concepts. This is where MJ’s weakness with on-product text hurts. You’ll often need to composite text in Photoshop or Figma after.
- Storyboarding and shot lists. Directors and DPs use Midjourney to rough out shot compositions before a real shoot.
Midjourney’s own team includes image generation, an Image Editor (still on V6.1), and a Video Model (also separate). The Editor is a separate tool with its own parameter set. The Video Model is only enabled on Pro and Mega and supports SD video on Standard, with HD video reserved for Pro and Mega.
Limitations & Gotchas
A few things nobody loves about Midjourney in 2026:
- No public API. This is the single biggest friction point for any team trying to build Midjourney into a product. Every “Midjourney API” on the market is an unofficial reverse-engineered Discord wrapper that violates the Terms of Service.
- No free tier. Free trials ended in March 2023. The cheapest entry is $10/month, or $8/month annually.
- Text rendering is still weak. Don’t trust Midjourney to spell “MORNING BREW” on a coffee bag mockup.
- No free license clarity for every use case. You own your outputs, but if you ask Midjourney to mimic a specific living artist or trademarked character, that’s on you.
- No offline mode. Everything is hosted. If Midjourney has an outage, you generate nothing.
- Rate limits on “unlimited” plans. Midjourney reserves the right to rate-limit even Mega subscribers to protect quality for everyone.
- Draft Mode and Omni Reference aren’t supported in V8.2 yet. These features still work on V8.1.
- Turbo mode is not supported in V8.1 or V8.2. You’ll have to use Fast or Relax.
- Public-by-default. If you don’t turn on Stealth (Pro/Mega), anything you post publicly can be remixed by other users.
- Legal overhang. Disney/Universal (June 11, 2025) and Warner Bros. Discovery (September 4, 2025) have active copyright suits against Midjourney. The outcomes could reshape training-data rules for the whole industry.
Privacy, Legal & Compliance
Midjourney’s Terms of Service, dated May 27, 2026, lay out a few things that matter for commercial users:
- You own the assets you create on a paid plan, to the fullest extent possible under applicable law. Your ownership is subject to third-party rights and the limits in this Agreement.
- The $1M revenue rule applies. Any company with more than $1,000,000 in annual gross revenue must be on Pro or Mega to own its outputs.
- Inputs and outputs are licensed back to Midjourney. By using the service, you grant Midjourney a perpetual, worldwide, sublicensable, royalty-free license to reproduce, prepare derivative works of, and distribute anything you input or generate. This license survives cancellation.
- Public content is remixable by default. Other Midjourney users can remix and reuse anything you post in a public channel or feed unless you have Stealth turned on.
- Arbitration, not court. Disputes are resolved through binding arbitration with the American Arbitration Association in Santa Clara County, California. You waive jury trial.
- DMCA takedowns are honored at takedowns@midjourney.com or via their web form.
- Minors (13+ allowed with parental consent). Midjourney tries to keep the service PG-13 but makes no guarantees about output suitability.
- Midjourney can suspend or ban you at any time for any reason.
On the litigation front:
- Disney and Universal filed suit June 11, 2025, calling Midjourney “a bottomless pit of plagiarism.”
- Warner Bros. Discovery filed suit September 4, 2025, focused on characters like Superman, Batman, Wonder Woman, Tweety, and Scooby-Doo. The complaint accuses Midjourney of offering “zero protection for copyright owners.”
- Earlier suits from Andersen v. Stability AI (January 13, 2023) and a follow-on November 2023 suit naming over 4,700 artists remain part of the broader training-data litigation landscape.
For most users, none of this changes day-to-day usage. But for brands and agencies, the legal picture is unsettled, and it’s worth a quick call to your counsel before using Midjourney outputs in high-stakes commercial contexts.
FAQ
Is Midjourney still the best AI image generator in 2026?
Yes, for raw aesthetic quality and creative direction polish, especially with V8.2’s bolder outputs and stronger personalization. But “best” depends on the job. For readable text in images, easy conversational prompting, and integrated workflows, ChatGPT (GPT Image) and Google’s Nano Banana Pro have closed the gap and now beat it for many casual and product tasks.
How much does Midjourney cost in July 2026?
Plans run from $10/month (Basic) to $30 (Standard), $60 (Pro), and $120 (Mega). Annual billing cuts each by 20%. Anything above Basic unlocks unlimited Relax-mode generations. Companies making more than $1,000,000 in annual gross revenue must subscribe to Pro or Mega to own their assets.
What’s new in Midjourney V8.2?
Released July 24, 2026. V8.2 is the new default. Midjourney describes it as more creative, bold, sophisticated, edgy, and fresh, with dramatically fewer random low-quality images and a much larger pool for personalization profiles. It does not change pricing, and it still has no public API.
Who owns the images I create with Midjourney?
You do, to the fullest extent possible under applicable law. The catch: companies making more than $1,000,000 in annual gross revenue must be on Pro or Mega to own their outputs. By default, anything you post publicly can be remixed by other Midjourney users unless you turn on Stealth Mode (Pro or Mega only).
Does Midjourney have a free plan or free trial?
No. Midjourney killed its free trial in March 2023 and has not brought it back. The cheapest entry point is the $10/month Basic plan, or $8/month on annual billing.
Is Midjourney better than ChatGPT image generation?
It depends. Midjourney wins on cinematic mood, lighting, and stylized concept art. ChatGPT’s image tools win on readable text in images, conversational prompting, and easy iteration inside a chat. Most professional designers I know use both.
What can I do with Midjourney Stealth Mode?
Stealth keeps your images and prompts out of the public Midjourney feed and out of remix rights for other users. It’s included on Pro and Mega plans only. For agencies and brands that can’t risk a draft leaking publicly, it’s a meaningful feature.
Why is Midjourney being sued by Disney and Universal?
Universal Pictures and The Walt Disney Company filed a copyright infringement suit on June 11, 2025, calling the service “a bottomless pit of plagiarism.” Warner Bros. Discovery followed with a similar suit on September 4, 2025. The cases target how Midjourney’s training data was sourced and how outputs can reproduce protected characters.
Can I use Midjourney images commercially?
Yes, if you’re on a paid plan and (if you’re a large company) on Pro or Mega. Midjourney grants ownership of your outputs under the Terms of Service, but you are responsible for making sure the output doesn’t infringe someone else’s rights. Don’t publish outputs that closely mimic a known living artist or trademarked character.
Did Midjourney buy Co-Star?
Yes. Midjourney announced its first-ever acquisition on July 23, 2026: Co-Star, the social astrology app with about 4.3 million monthly active users and tens of millions of registered accounts. Co-Star founder Banu Guler is joining Midjourney as Chief Design Officer, while Co-Star continues operating independently under her direction.
Final Verdict
Midjourney in July 2026 is still the highest-aesthetic, most opinionated AI image generator on the market, and V8.2 makes it a little better than V8.1 at exactly the things Midjourney already did well. The pricing is unchanged, the workflow is unchanged, and there is still no public API. If you already have a Midjourney habit, the V8.2 upgrade is a free improvement. Take it.
What changed in 2026 isn’t the model. It’s the company. The Co-Star acquisition, Midjourney’s first ever, signals that David Holz is no longer satisfied being a Discord-first model lab. With Banu Guler as Chief Design Officer and a consumer app team in tow, the next big Midjourney release may not be a new model version at all. It might be a standalone Midjourney app.
For now, treat Midjourney V8.2 as what it’s always been: the best-looking image generator you can subscribe to, paired with the manual workflow you’ve come to know. Use it for what it’s great at. Use ChatGPT or Nano Banana Pro for what they’re great at. And if your legal team is nervous about training-data litigation, run a quick check before putting Midjourney imagery on a billboard.
Jobs to be done
Use cases
- 01Concept art and moodboards for campaigns
- 02Brand visual exploration and style tests
- 03Product mock imagination and packaging concepts
- 04Social and editorial imagery ideation
- 05Creative direction references for human designers
Capabilities
Features
Generation
- Text-to-image generation
- Image variations and remix-style iteration
- Upscaling and refinement tools
Creative control
- Style and parameter controls
- Reference-driven prompting workflows
Collaboration
- Shared creative exploration in community or team contexts
Tradeoffs
Strengths & limitations
Strengths
- Strong visual quality and aesthetic range for many creative styles
- Excellent for rapid ideation and moodboard creation
- Mature iterative workflow for exploring variations
- Widely used reference point in creative AI discussions
Limitations
- Not a complete replacement for layout, brand systems, or multi-page design tools
- Prompt skill and art direction still matter for consistent brand results
- Subscription required; free access has not been the long-term default model
- Commercial usage rights and content policies must be reviewed for each use case
Evidence
Sources
- 01
- 02
Related
Related guides & workflows

Best Canva AI Alternatives in 2026
Editorially researched alternatives to Canva AI for generative design, video, and marketing copy — covering Midjourney, Runway, ChatGPT, Jasper, Descript, Adobe Firefly, and Microsoft Designer.

Best Midjourney Alternatives in 2026
Looking past Midjourney for image generation? These 2026 alternatives cover the real image-model peers — DALL·E, Stable Diffusion, Adobe Firefly, Ideogram, Leonardo AI, FLUX, and Recraft — without invented scores.

Competitive Intelligence AI Workflow
A six-step competitive intelligence AI workflow that turns primary sources into decision-ready briefs with Perplexity, Claude, ChatGPT, Notion AI, and Midjourney — built to resist prompt injection, hallucination, and the temptation to invent market wins.

AI Content Marketing Pipeline
A human-led AI content marketing pipeline for turning audience evidence into cited, on-brand campaigns and measurable demand.

Design-to-Campaign AI Workflow
A practical multi-step workflow for turning visual concepts into full campaign packages—creative direction, copy variants, motion assets, and publish-ready channel kits—using real AI design and marketing tools with human brand and claims review.

Best AI Design Tools for Marketing & Creative Teams in 2026
A firsthand, evidence-checked shortlist of the best AI design tools marketing and creative teams actually use in July 2026 — Canva AI, Midjourney, Runway, ChatGPT, Jasper, and Descript — with capability descriptions, pricing links, and primary sources for every product fact.

Best AI Marketing Tools in 2026 (Tested, Brand-Verified Picks)
A first-person collection post of AI marketing tools for content, creative, and go-to-market work—selected for real channel fit, cross-verified against brand pages, independent analyst reports, and July 2026 release notes. No paid placement, no invented scores.

Best AI Video Tools
A research-led guide to the best AI video tools for generative filmmaking, video transformation, avatars, localization, transcript editing, effects, pre-production, and audio finishing in 2026.
Catalog
Related products
Canva AI
Editorially ResearchedAI design generation and Magic Studio tools inside Canva's accessible design platform.
Stands out · AI-assisted design inside a mainstream template-driven creative platform with brand systems, Magic Studio tools, native AI-assistant connectors and enterprise-grade indemnification.
Runway
Editorially ResearchedAI creative suite for generative video, image, audio, and now world-model infrastructure for marketing, film, and enterprise production.
Stands out · An AI creative platform that pairs frontier video models (Gen-4.5, Aleph 2.0) with a true production workspace, the Runway Agent, and a developer platform with a Model Router.
ChatGPT
Editorially ResearchedOpenAI's general-purpose AI assistant for writing, analysis, coding help, and everyday knowledge work.
Stands out · A mainstream, general-purpose AI assistant with the broadest multimodal feature surface and one of the largest everyday user bases for conversational AI.
Descript
Editorially ResearchedAI video and podcast editor that cuts by editing the transcript, with 2026-era Underlord agent, Studio Sound, and 4K remote Rooms.
Stands out · A creator-first editor where the transcript is the primary interface, now expanded into an agentic AI workspace with Underlord, 4K remote Rooms, and regenerative video inpainting.
Related
Compare this product

Midjourney vs Canva AI
Choose Midjourney when distinctive generative imagery is the bottleneck. Choose Canva AI when finishing, formatting, and shipping campaigns at volume is the bottleneck. Many teams use both.

Midjourney vs Runway
Choose Midjourney when the deliverable is a still image, moodboard, or visual concept. Choose Runway when the deliverable moves—video, real-time avatars, world models, or motion prototypes for film and marketing.

Midjourney vs ChatGPT
Choose Midjourney for image-first aesthetic work, moodboards, and concept art. Choose ChatGPT when images are one part of a broader drafting, research, or agent workflow.
Next step
Explore Midjourney
Visit the official site, or browse the full catalog for similar tools.