Midjourney
AI image generator for concept art, previs, boards and client-safe stills, with private Stealth mode
Midjourney generates images from text and reference images on its web app and Discord. The current default is V8.2 (released July 24, 2026), which renders native 2048x2048 HD images. A V8.2 Edit model in public test since August 27, 2026 adds instruction edits, inpainting and outpainting.
- 106 creators using
- 2 news items tracked
- 2022 first launched
- Paid pricing
- Easy learning curve
What Midjourney can do
- Infinite canvas Partial
- AI agent No
- Node / flow builder No
- Public API No
- MCP server No
9 of 26 capabilities confirmed
| Capability | Supported | Details |
|---|---|---|
| Text to image | Yes | V8.2 default; standard 1024x1024, HD 2048x2048 |
| Image editing | Yes | V8.2 Edit model (instruction edits, inpainting, outpainting), public test since Aug 27, 2026 |
| Text to video | Not confirmed | |
| Image to video | Yes | V1 Video (June 2025), four 5-second clips per job |
| Video to video | No | |
| Native audio in video | Not confirmed | |
| Lip-sync | No | |
| Voice / TTS | No | |
| Music & SFX | No | |
| Upscaling | Yes | Upscale and zoom in the web app; native HD 2K |
| Character consistency | Partial | Up to 4 references in V8.2 Edit model |
| Camera control | Not confirmed | |
| Custom training | Partial | Personalization profile built from your ratings, not a LoRA or custom model |
| Third-party models | No |
| Capability | Supported | Details |
|---|---|---|
| Infinite canvas | Partial | Moodboards and folders in the web app |
| Node / flow builder | No | |
| AI agent | No | |
| Chat assistant | No | |
| Team collaboration | Partial | Shared Discord server and public gallery |
| Capability | Supported | Details |
|---|---|---|
| Public API | No | No official public API; terms prohibit automated access |
| MCP server | No | |
| Mobile app | Not confirmed | |
| Desktop / local | No | |
| Open source | No |
| Capability | Supported | Details |
|---|---|---|
| Free tier | No | No free trial |
| Commercial use | Yes | All paid plans; companies over $1M gross annual revenue need Pro or Mega |
Yes Partial No Not confirmed by the vendor
What makes Midjourney different
- 01
Stealth mode on Pro and Mega keeps confidential client and NDA work out of the public gallery, which the tools above do not offer on an equal basis.
- 02
Consistent, art-directed default look across prompts with the V8.2 personalization profile, which is trained on your ratings rather than a LoRA you build.
- 03
Native 2048x2048 HD output without a separate upscale step, on every paid tier.
- 04
Editing now covers instruction edits, inpainting, outpainting and up to four references in one model, in the web app lightbox and on Discord.
- 05
Paid plans include commercial rights, so a small production can use the images without a separate licence.
Latest from Midjourney
- Apr 30 Version 8.1 Enhances Image Sharpness and Style Reference Fidelity Midjourney V8.1 introduces improved image quality and better visual consistency for Style References and Moodboards across Discord and web platforms.
- Apr 27 Community Ranking Opens for Native 2K High-Resolution Image Tuning Midjourney is collecting user feedback on raw 2K resolution images to refine the aesthetic quality of its upcoming high-definition model updates.
About Midjourney
Midjourney is still the default look for concept art, storyboard frames and mood boards. The image model moved quickly through 2026: V8 Alpha in March, V8.1 as the default on June 10, and V8.2 as the default on July 24. V8.2 is pitched as an aesthetics and personalization release, with native 2K HD output and fewer random low-quality results. Standard renders are a 1024x1024 base, and HD is an option in the settings panel or with --hd.
The biggest change in 2026 is editing. The V8.2 Edit model, announced August 27, 2026, takes a plain-language instruction on an uploaded image, changes selected areas (inpainting), extends the frame (outpainting), and accepts up to four reference images at once. Midjourney says this replaces Omni Reference for V8 work. Midjourney calls it an early test, so expect edge cases. Video is still the V1 image-to-video model from June 2025, which makes four five-second clips per job.
For film work, Midjourney suits previs, look development and key art where a consistent frame matters more than shot-level control. Paid plans include commercial rights, but companies above $1M in gross annual revenue need Pro or Mega. Pro adds Stealth mode, which keeps client work out of the public gallery, and Relax video. There is no free trial.
What's new in Midjourney
-
V8.2 Edit model opens for testing
Instruction editing, multi-image input (up to 4 references), brush inpainting and outpainting in one model, replacing Omni Reference for V8 work.
Source: updates.midjourney.com ↗ -
V8.2 becomes the default image model
Released July 24, 2026 as an aesthetics and personalization update with native 2048x2048 HD output.
Source: cometapi.com ↗ -
V8.1 becomes the default image model
Default from June 10, 2026, with faster standard rendering and HD available through the settings panel or --hd.
Source: atlascloud.ai ↗
Models available in Midjourney
-
Midjourney V1 Video Midjourney Image-to-video, four 5-second clips per job; launched June 2025
-
Key Features
- V8.2 default image model with native 2048x2048 HD output
- V8.2 Edit model (test): instruction edits, inpainting and outpainting
- Up to four reference images in one generation
- Personalization profile built from your ratings, plus moodboards and srefs
- Image-to-video (V1) for four 5-second clips per job
- Stealth mode for private generations on Pro and Mega
- Relax mode for slower, unlimited generation on eligible plans
- Remix, vary, zoom and upscale in the web app and on Discord
When to reach for it — and when to skip
Reach for it when…
- Strongest default aesthetic for key art, mood boards and concept frames
- Native HD output without an upscale step
- Stealth mode and commercial rights on paid plans
- Fast standard renders
- Edit model covers inpainting and outpainting in one place
Skip it when…
- No free trial
- Video is still the 2025 V1 image-to-video model, with no text-to-video or audio
- No official API, so pipelines have to go through the app or Discord
- Edit model is a test; expect edge cases while it is in public beta
Best For
✓ Ideal for
- Concept art, key art and look development for features and series
- Storyboard frames and mood boards where the look matters more than shot control
- Client-confidential previs on Pro or Mega using Stealth mode
- Poster, festival and marketing stills that need HD output
- Iterating a locked frame with inpainting and outpainting instead of regenerating it
✗ Not built for
- Shot-by-shot video generation with camera control or dialogue
- Teams that need an API, MCP server or automated pipeline
- Budget users who want a free tier
Working Tips from Filmmakers Using Midjourney
- 01 Relax mode is slower but does not use Fast GPU time, so it suits exploration.
- 02 Use the V8.2 Edit model with an uploaded still to change one area (inpaint) so the rest of a locked frame stays the same across a storyboard.
- 03 Stealth mode is on Pro and Mega only. Turn it on before you generate client boards, not after.
Midjourney pricing in 2026
| Plan | Price | What you get |
|---|---|---|
| Basic | $10 monthly ($8/mo billed annually) |
|
| Standard | $30 monthly ($24/mo billed annually) |
|
| Pro | $60 monthly ($48/mo billed annually) |
|
| Mega | $120 monthly ($96/mo billed annually) |
|
Prices checked against Midjourney's pricing page on 9 Oct 2026. Vendors change plans often, so confirm before you buy.
Midjourney vs Flux vs Leonardo.ai vs Adobe Firefly
| Midjourney | Flux | Leonardo.ai | Adobe Firefly | |
|---|---|---|---|---|
| Price | From $10/mo ($8/mo billed annually); 20% off on annual plans | Pay per image on the API, from $0.014 (FLUX.2 klein 4B). Open weights are free for non-commercial use; commercial open-weights licences are quoted by sales. | See the vendor's pricing page | Free daily generations, or Standard from $9.99/mo (2,000 credits); Pro Plus and Premium have a 30% first-year promo ending Oct 21 |
| Infinite canvas | Partial | No | Yes | Yes |
| Node / flow builder | No | No | Partial | Not confirmed |
| AI agent | No | No | Not confirmed | Yes |
| Chat assistant | No | No | Not confirmed | Yes |
| Team collaboration | Partial | Not confirmed | Partial | Yes |
| Text to image | Yes | Yes | Yes | Yes |
| Image editing | Yes | Yes | Yes | Yes |
| Text to video | Not confirmed | Yes | Yes | Yes |
| Image to video | Yes | Yes | Yes | Yes |
| Video to video | No | Partial | Not confirmed | Yes |
| Native audio in video | Not confirmed | Yes | Partial | Not confirmed |
| Lip-sync | No | Yes | Not confirmed | Not confirmed |
| Voice / TTS | No | Partial | Not confirmed | Yes |
| Music & SFX | No | Partial | Not confirmed | Yes |
| Upscaling | Yes | Yes | Partial | Yes |
| Character consistency | Partial | Partial | Partial | Partial |
| Camera control | Not confirmed | Partial | Partial | Not confirmed |
| Custom training | Partial | Yes | Yes | Yes |
| Third-party models | No | No | Yes | Yes |
| Public API | No | Yes | Yes | Yes |
| MCP server | No | Yes | Not confirmed | Not confirmed |
| Mobile app | Not confirmed | Not confirmed | Yes | Not confirmed |
| Desktop / local | No | Yes | No | Partial |
| Open source | No | Partial | No | No |
| Free tier | No | Partial | Yes | Yes |
| Commercial use | Yes | Partial | Yes | Yes |
Each column comes from that tool's own researched page. "Not confirmed" means the vendor doesn't say.
Open the full Midjourney vs Flux vs Leonardo.ai vs Adobe Firefly comparison →Use Cases
Integrations
Tags
Sources
Facts on this page were checked against Midjourney's own pricing, product and changelog pages on , then cross-checked with independent coverage where vendor pages were silent. Anything we couldn't confirm is marked "Not confirmed" rather than guessed.
- Pricing page docs.midjourney.com
- Midjourney update: V8 edit model (Aug 27, 2026) updates.midjourney.com
- CometAPI: Midjourney V8.2 facts (dated Oct 1, 2026) cometapi.com
- Atlas Cloud: Midjourney V8.1 (June 2026 default date) atlascloud.ai
- eesel: Midjourney pricing in 2026 (last edited June 5, 2026) eesel.ai
- AI Tools Dev Pro: Midjourney 2026 guide (accurate as of Jan 2026) aitoolsdevpro.com
- Adobe Firefly plans (comparison column) adobe.com
Spotted something out of date? Tell us and we'll re-check it.
Discussion
No comments yet — be the first.