⚡️ The Midjourney V6 alpha version is available on @GPT4Telegrambot

Midjourney has unveiled access to their new V6 model — it follows prompts more accurately, boasts even more aesthetic appeal, and can draw text.

⭐️ What's new?
— V6 better understands your prompt and pays attention to details;
— Minor text drawing ability;
— Some parameters are supported: --ar, --chaos, --weird, --tile, --style raw (more will come over);
— The results are beautiful 🙂

▶️ How to start using it?
— On @GPT4Telegrambot, buy Midjourney in the /premium section;
— To generate an image, use the command /imagine followed by a text description;
— Type --v 6 at the end of your prompt to use the new model.

ℹ️ Style and prompting for V6:
— To draw text, write it in "quotations";
— Avoid 'junk' like "photorealistic, 4k, 8k". If you want something more photographic, use the --style raw parameter;
— Be explicit about what you want: the fewer words you use, the more significant each one becomes.

🗣️ Example: /imagine a photo of the text "Hello World" written with a marker on a sticky note --v 6

Find more examples above 🔼

Midjourney has trained the new model from scratch on their AI superclusters. It's been in the works for 9 months. It's still in the alpha stage, so changes and further refinements will occur.

The company is confident that V6 takes image generation to new heights. It can generate much more realistic imagery than anything released before. Our editorial team has found it impressive. Give it a try and share your impressions.

#Midjourney @hiaimediaen