Midjourney V8.2 Text to Image: API, Specs & Pricing

The short answer
Midjourney V8.2 Text to Image is now callable through a model API, returning four images per prompt with optional native 2K HD, a style reference, and stylize, chaos and weird controls. V8.2 has been Midjourney's default model on midjourney.com since July 24, 2026, and sibling V8.2 tools (Blend, Image to Image, Remove Background, Style Transfer) run in the AdGraam Playground from $0.086 per run.
On this page (10 sections)
Midjourney V8.2 Text to Image is now callable on a model API, according to public model registry metadata seen on November 11, 2026. The model generates four images from a text prompt, with optional native 2K HD, a style reference, and aspect ratio, stylize, chaos and weird controls.
The V8.2 model has been around for months. It has been Midjourney's default since July 24, 2026. What changed is access: reviews as recently as October said Midjourney had no public API.
What is Midjourney V8.2 Text to Image?
Midjourney V8.2 Text to Image is an image generation model from Midjourney, an independent San Francisco research lab founded by David Holz, that turns a written prompt into a grid of four images. It is the text-prompt entry point of the V8.2 family, which also includes editing and blending tools.
Midjourney describes the V8.2 look as "more creative, bold, sophisticated, edgy and fresh," according to a Unite.AI review. The version focuses on aesthetics, image quality and personalization, per the Founderz overview, which cites Midjourney's documentation.
For ad teams, the news is where the model can now be used. Midjourney's image quality has long been the draw, but access was limited to its website and Discord. An API route lets the model sit inside automated creative pipelines.
Midjourney V8.2 release date and availability
V8.2 replaced V8.1 as the default model on midjourney.com on July 24, 2026, according to Midjourney's version documentation as cited by Founderz, Unite.AI, AI Worth It and Gradually. The Text to Image model's appearance on a model API was first seen November 11, 2026.
The version history, as reported by reviewers:
- V7 (April 2025): Draft Mode and Omni Reference.
- V8.0: a large infrastructure rebuild, with community pushback on its look.
- V8.1 (alpha April 14, 2026, rollout April 30, 2026): native 2K and much faster rendering, per Fello AI.
- V8.2 (default since July 24, 2026): aesthetics and personalization.
Midjourney has not published its own API announcement in the available material. A Fello AI review updated October 15 stated there was still no public API. The API listing comes from registry metadata, not from Midjourney, so check the terms of whichever provider you use.
Midjourney V8.2 Text to Image key features and specs
V8.2 Text to Image returns four images per prompt, supports optional native 2K HD output, accepts a style reference, and exposes aspect ratio, stylize, chaos and weird controls. Exact ranges for those controls are not listed for this endpoint. The sibling V8.2 tools in the AdGraam Playground list the ranges noted below.
Control dials and sliders representing stylize, chaos and weird parameters
| Spec | Details |
|---|---|
| Model | Midjourney V8.2 Text to Image |
| Input | Text prompt, optional style reference |
| Output | 4 images per run |
| Resolution | Native 2K HD (optional); V8.1 HD was 2048 x 2048 |
| Aspect ratios (V8.2 Blend and Image to Image) | 1:1, 9:16, 16:9, 4:3, 3:4, 2:3, 3:2, 9:21 |
| Stylize, chaos, weird | Controls supported; ranges for Text to Image not announced (Blend and Image to Image list 0 to 1000, 0 to 100 and 0 to 3000) |
| Audio / video | Not applicable (image model) |
| API price (Text to Image) | Not announced |
| Status | Callable on a model API |
The aspect ratio list matters for paid social. 9:16 covers Reels and Stories, 1:1 and 3:4 cover feed placements, and 16:9 covers YouTube and landscape placements.
Reviewers also comment on output quality. Unite.AI says V8.2 "finally got short text right" and kept a character consistent across scenes in its test, though it also reports a backwards coat that Thinking Mode and the Editor did not fix. AI Worth It still points to a text specialist such as Ideogram 4.0 for logos, posters and packaging.
Midjourney V8.2 pricing and free access
Midjourney has no free plan. Its own plans run $10 per month (Basic), $30 (Standard), $60 (Pro) and $120 (Mega), with 20% off yearly billing, per AI Worth It's prices checked November 5, 2026. API pricing for Text to Image is not announced in the available material.
Other details reviewers report:
- Basic includes 3.3 hours of Fast GPU time, and HD burns through it quickly, per Unite.AI.
- Unlimited Relax-mode images come with Standard and higher plans.
- Images are public by default. Stealth Mode needs the $60 Pro plan.
- Commercial use is included with every plan, per Unite.AI.
In the AdGraam Playground, four V8.2 tools are live with per-run pricing:
| Tool | Output | Price |
|---|---|---|
| V8.2 Blend (2 to 5 images in) | 4 images | $0.086 per run |
| V8.2 Image to Image | 4 images | $0.086 per run |
| V8.2 Style Transfer | 4 images | $0.129 per run |
| V8.2 Remove Background | 1 transparent image | $0.086 per image |
Pricing for the Text to Image endpoint is not listed. For per-run details on the related tools, see the breakdowns for Blend, Image to Image and Style Transfer.
Midjourney V8.2 vs competing AI image models
Midjourney V8.2 leads on look and artistic control, according to AI Worth It, but benchmarks are mixed. In Gradually's GraduallyBench, Midjourney V8.1 ranked 67 of 108 and Muse Image led with 90.1 points, while GPT Image 2.5 Sunburst won on LMArena.
Those results cover V8.1, not V8.2, and benchmarks reward different things than ad performance. A comparison from reviewer material:
| Model | Free option | Strength per reviewers |
|---|---|---|
| Midjourney V8.2 | No | Aesthetics, style control, personalization |
| ChatGPT Images 2.5 | Limited free | Images inside a wider AI workflow |
| Nano Banana 2 | Limited free | Casual image generation |
| Ideogram 4.0 | Not stated | Accurate text in images |
For a newer text-focused option, see AdGraam's coverage of Ideogram V4.5. For another high resolution contender, see FLUX 3 Image.
How to use Midjourney V8.2 for AI video and static ads
V8.2 Text to Image fits the concepting end of a creative workflow: generating four distinct directions per prompt for hero images, backgrounds, lifestyle scenes and style explorations. Text-heavy layouts and label accuracy still need another model or a manual pass.
A desk of printed product ad mockups arranged in a grid for creative testing
Where it fits:
- Static ad concepts. Run one prompt per angle and test four visuals at once. Use 3:4 for feed or 9:16 for Stories. Our guide to making static ads with AI covers the brief structure.
- Product scenes. Generate the environment, then use Blend or Image to Image to place your real product shot into the style. Remove Background gives a transparent cutout for compositing.
- Style consistency. Use a style reference so a campaign set looks like one family.
- Video b-roll. Midjourney can animate stills, but Unite.AI reports the clips are silent, limited to 720p image to video, and warp small text. Do not rely on it for shots where a label must stay legible.
If you want to compare Midjourney against other image models on the same prompt, the AdGraam Playground runs 200+ models in one place. To go from a product URL to finished static ads, use the Creative Generator.
One caution: Midjourney's output is strongest as a creative starting point. Reviewers flag copyright lawsuits from Disney, Universal and Warner Bros. Discovery against Midjourney, and legal teams at larger brands may want to review the terms before using outputs in paid media.
What to watch next
Three things are unresolved: whether Midjourney confirms an official API, what per-run pricing applies to Text to Image, and how V8.2's Edit Model and the Thinking Mode Midjourney began testing in October 2026 reach API users.
Watch for a Midjourney announcement or documentation update that confirms the API terms, since the current listing comes from registry metadata. Also watch whether the 2K HD option is priced separately from standard runs, as it effectively is on the Midjourney site, where HD uses Fast hours faster.
Until Text to Image pricing is published, the Playground rates for the sibling V8.2 tools ($0.086 to $0.129 per run) are the only V8.2 price points available.
Key takeaways
- Midjourney V8.2 Text to Image is now listed as callable on a model API, generating four images per text prompt.
- Native 2K HD is optional, and the model supports a style reference plus aspect ratio, stylize, chaos and weird controls.
- V8.2 became Midjourney's default model on July 24, 2026, replacing V8.1, according to Midjourney's version documentation as cited by reviewers.
- Midjourney plans on its own site run $10 to $120 per month, with no free plan.
- Four V8.2 sibling tools run in the AdGraam Playground: Blend and Image to Image at $0.086 per run of 4 images, Style Transfer at $0.129, Remove Background at $0.086 per image.
- A per-run API price for Text to Image itself is not announced in the available material.
Compare 200+ AI video and image models side by side
Run Kling, Veo, Sora, Seedance, Flux, Ideogram and Recraft in one Playground, or turn any product URL into ready-to-launch ads.
Frequently asked
When did Midjourney V8.2 come out?
V8.2 became the default model on midjourney.com on July 24, 2026, replacing V8.1, according to Midjourney's version documentation cited by several reviewers. The Text to Image model appeared as callable on a model API on November 11, 2026.
How much does Midjourney V8.2 Text to Image cost?
On midjourney.com, plans run $10 per month (Basic) to $120 per month (Mega), per reviews checked in October and November 2026. API pricing for Text to Image specifically is not announced. In the AdGraam Playground, related V8.2 tools cost $0.086 to $0.129 per run.
Is Midjourney V8.2 free to try?
Not on Midjourney's own site. Reviewers report no free plan and no trial on midjourney.com or Discord. Access through an API or a multi-model platform is the alternative route.
Does Midjourney V8.2 have an API?
Reviews published in October 2026 said Midjourney had no public API. A public model registry now lists V8.2 Text to Image as callable on a model API, and Midjourney has not published its own API announcement in the available material.
Can you use Midjourney V8.2 for commercial ads?
Reviewer Unite.AI states commercial use is included with every Midjourney plan. Images are public by default, and Stealth Mode requires the $60 per month Pro plan. Check the license terms of whichever access route you use.
What is new in V8.2 compared with V8.1?
Midjourney describes V8.2 as focused on aesthetics, image quality and personalization. Reviewers also credit it with better short text rendering, and it introduced an Edit Model that replaces Omni Reference, Character Reference and Retexture.
Sources
- What is Midjourney AI? How the AI image generator ...founderz.com
- Midjourney V8.1 Review: HD by Default, 5x Faster (2026)felloai.com
- Midjourney Review: Can Its AI Videos Get Text Right?unite.ai
- Midjourney Review 2026: Is It Still the Best AI Image Generator?aiworthit.com
- The Best Midjourney Alternatives in 2026 (Free Options Too)gradually.ai
More AI model news & creative guides

Midjourney V8.2 Remove Background: $0.086 per Image
Midjourney V8.2 Remove Background returns one transparent cutout for $0.086 per image in the AdGraam Playground. Get specs, ad workflow tips and access details.

BouncyBohr: Leaked Google Video Model, 4K Claims
BouncyBohr is a rumored Google video model with 4K and 10-second extensions, per a leaker. Get the spec table, ad workflow breakdown and access details.

Midjourney V8.2 Image to Video: 4 Clips, 720p
Midjourney V8.2 Image to Video turns one image into four 5-second clips at 480p or 720p. Get the spec table, ad workflow breakdown and access details.
Best AI Ad Generators for D2C Brands in 2026
Ten AI ad generators compared for D2C and e-commerce brands: what each makes, real starting prices, free options, and which suits Indian brands paying in rupees.
Break Even ROAS for COD Orders: How RTO Changes the Math
How to work out break even ROAS when a share of COD orders comes back as RTO, with worked rupee examples, a prepaid and COD mix, and why Ads Manager ROAS overstates COD.
What Indian D2C Brands Run on Meta: 2,790 Ads Analyzed
We studied 2,790 Facebook and Instagram ads from 14 Indian brands and their 747 longest-running ads: formats, angles, offers, languages and CTAs that last.
Your move
See your product as a winning ad
Turn a product URL into static ads and UGC-style videos, rebuild the competitor ads that already work, and read your Meta data to know what to make next.
AdGraam News DeskAI video and image news
The AdGraam News Desk (part of AdGraam, also searched as Adgram or Adgram AI) tracks AI image and video model launches, leaks and rumors from labs, leakers and creators, and checks each claim against its sources. Unconfirmed stories say so, and each page is updated as its story develops.