Midjourney V6 alpha test started

Midjourney V6 alpha test started

Developers of Midjourney announced about the start of the V6 alpha test, which will better understand the prompts. In addition, the neural network has learned to display text more accurately without blurring and supports the option of fine-tuning textures or lighting.

To activate Midjourney V6, you need to send a direct message to the server or bot in Discord, enter the command / settings, and then select V6 from the drop-down menu. Alternatively, the version can be run by typing “-v 6” at the end of the prompt.

David Holtz, founder of Midjourney, mentioned that when asking, avoid phrases like “award-winning, photo-realistic, 4K, 8K” and be specific. If you want more realistic results, it’s better to use –style raw by default.

The prompt structure below will help you get the best results.

1. Style:

  • Purpose: Provides a certain aesthetic or artistic direction.

  • Details to include: preferred style or era.

2. Topic:

  • Target: Defines the main focus of the image.

  • Details to include: Characteristics of the main object (such as a person or animal), including appearance, colors and unique features.

3. Settings:

  • Purpose: Sets the environment or context for the item.

  • Details to include: location (indoor, outdoor, imaginary), environmental elements (nature, city), time of day, and weather conditions.

4. Composition:

  • Purpose: Defines how the object and elements are designed and viewed.

  • Details to include: viewpoint (close-up, wide, bird’s-eye), angle, and special framing options.

5. Lighting:

  • Purpose: sets the mood and visual tone of the image.

  • Details to include: type of lighting (bright, dim, natural), mood (fun, mysterious) and atmospheric effects.

6. Additional information:

  • Purpose: Adds complexity and depth to the image.

  • Details to include: secondary objects, characters, animals and their interaction or location in relation to the main object.

Related posts