Midjourney launches its new image model V7 AI that can process text promotions better

Midjourney have released the alpha version of V7, which is said to be "new" AI image generation model and is much more inherent in processing your text prompts. The quality of the image of its output is noticeably higher, says the Midjourney team and can create better textures, bodies and hands. AI image generators usually fight the creation of accurate hands images, but based on photos posted by some users of their social, V7 is able to spit out some photorealistic images of human hands.

The new model comes with a feature called "Draft mode," which can present images with half the speed that the program usually accepts. Its results are more coarse and less detailed, but it will cost half of what is worth the standard generation. Midjourney says the draft mode is the best way to repeat ideas. You can use it, say, if you collaborate with someone and just spit ideas with them – he works with a voice, so you don’t even have to write your suggestions to each other – or if you are not sure what kind of vibration you are getting on. If you like an image that Midjourney creates in draft mode, you can click on "Improve" or "vary" on it to convert it with full quality.

At startup, the V7 will have turbo and relaxation regimens for standard depiction, with the former worth twice as much as the normal speed of the V6. Midjourney needs more time to optimize its standard V7 speed mode, but this will make the option available in the future. The new mode also lacks more options, including enlargement, editing and retecting, which will now return to the V6 model of the program. Midjourney promises to introduce new updates for the model every week or two over the next two months.

In order to test the alpha version of the V7, you will first need to unlock your personalization account. Midjourney describes personalization as "Style assistant for your images," As he teaches AI your visual preferences. You will need to rank at least 200 pairs of images to create a V7 global customization profile and test the model. Personalization is switched on by default for the V7, but you will be able to turn it off if you want.

This article originally appeared on Engadget at https://www.engadget.com/ai/midjourney-launches-its-b7-V7-ii-Image-Model-Text-txt-prompts-Better-134545

 

Report

Thank you for Report

Game / Application Name

Your Email: *

Issue: * I can't download the APK file I can't install the APK file The file is not compatible File does not exist Update request Copyright Infringment Spam Invalid Contents Broken Links Others

Submit