TellWorldAI World Archive

Home›Models

Model

Midjourney V2

Midjourney V2 was the second model version of Midjourney's image generation model, released on 12 April 2022 and the default model until V3 arrived that July. Midjourney later described it as colorful and painterly but low in coherency.

Modality Text, Image Era Large models

License
Proprietary
Access
App
Release date
Developer
Midjourney

What Version 2 was

Midjourney V2 was the second model version of Midjourney's image generation model, released on 12 April 2022 [1]. In systems of this kind a version number marks a distinct generation of the model: the same written prompt given to two versions can come back as visibly different pictures, which is why users who want a particular look keep track of which version produced it. When V2 was current, the route to the model was the Midjourney Discord bot, which is how the service was used before the open beta [1]. The model was therefore reached inside a chat application, through the bot, in that period.

The default model

Midjourney's documentation records that Version 2 was the default model from April 2022 to July 2022 [2]. "Default" means the version that answers a prompt when the user does not ask for a particular one: during those months, a prompt that named no version was handled by V2. The window Midjourney records for it begins with its April release and ends with the arrival of its successor in July, so V2's time as the default covers the gap between the two releases.

The traits Midjourney records

The same documentation describes Version 2 as creative, colorful and painterly, with low coherency [2]. Each word points at something a viewer can see. "Painterly" describes images that look as though they were made with paint rather than photographed — visible marks, blended colour, a picture that reads as an illustration. "Colorful" and "creative" describe the palette and the willingness of the model to invent compositions rather than produce something plain. "Low coherency" is the limiting trait, and it is worth spelling out: coherency is the degree to which the parts of a picture agree with one another — whether limbs and hands are put together plausibly, whether the perspective holds, whether an object keeps the same shape from one side of the frame to the other. A model with low coherency can produce a striking overall composition while details inside it come out garbled or inconsistent. The combination Midjourney records — strong colour and painterly surfaces on one side, weak internal consistency on the other — is a specific description of what V2's pictures tended to look like.

Handover to V3

The successor, V3, was released on 25 July 2022 [1]. According to the same documentation, Version 3 was the default model from July 2022 to November 2022 [2]. The two records fit together: V2's documented run as the default ends where V3's begins, and the month of the handover is the same in both.

Where the description comes from

Midjourney keeps an account of its older versions on a Legacy Features documentation page [2], and that page is the source for the characterisation above. It is a particular kind of evidence: the maker's own later summary of what each version was like, rather than a review written while the version was current. Read that way, V2 is the version associated with the stretch between the Discord-bot era before the open beta [1] and the July 2022 switch to V3.

Version history

8
  1. 2022
    • Feb 2022 Midjourney V1

      Midjourney V1 was Midjourney's first image-generation model, released in February 2022 and used through a bot on the company's Discord server. It produced abstract, painterly images and was replaced by V2 and V3 within months.

    • Apr 12, 2022 Midjourney V2 · This page

      Midjourney V2 was the second model version of Midjourney's image generation model, released on 12 April 2022 and the default model until V3 arrived that July. Midjourney later described it as colorful and painterly but low in coherency.

    • Jul 25, 2022 Midjourney V3

      Midjourney V3, released on 25 July 2022, was the third model version of Midjourney; it raised image quality sharply by learning from how users behaved, and it produced the image that won a Colorado State Fair digital-art prize.

    • Nov 5, 2022 Midjourney V4

      Midjourney V4 is the fourth-generation image generation model released in alpha on 5 November 2022. It was the first Midjourney model trained on Google TPUs, and reviewers described it as a large step up from V3.

  2. 2023
    • Mar 15, 2023 Midjourney V5

      Midjourney V5 is a version of the Midjourney text-to-image model, released in alpha on 15 March 2023 and refined through the 5.x line before being succeeded by V6 in December 2023.

    • Dec 21, 2023 Midjourney V6

      Midjourney V6 is a version of Midjourney's image generation model, released in alpha on 21 December 2023 and trained from scratch over nine months; its 6.1 update followed in July 2024.

  3. 2025
    • Apr 4, 2025 Midjourney V7

      Midjourney V7 is an image generation model released as an alpha to Midjourney's community on 4 April 2025, the first of the company's models to have personalization on by default and the first to add a cheap, fast Draft Mode.

  4. 2026
    • Mar 17, 2026 Midjourney V8

      Midjourney V8, released as an alpha on 17 March 2026, generates images about five times faster than earlier versions, adds a native 2K mode, and improves rendering of text inside images.

Developed by1

Midjourney

Successor of1

Midjourney V1

Succeeded by1

Midjourney V3

Notes and sources

  1. Midjourneyhttps://en.wikipedia.org/wiki/Midjourney
  2. Legacy Featureshttps://docs.midjourney.com/hc/en-us/articles/33329788681101-Legacy-Features

Cite this page: Midjourney V2. TellWorld, https://www.tellworld.net/models/midjourney-v2. Updated 2026-11-08