Krea2 V2.0

What's new ?

• Upload up to 4 images as reference

• LLM chatbot added (same as Minimax version)

• Fields for Lora triggers words in generator and modificator

• "Vision" node can be bypass to save some VRAM when not using image analysis

Minimax H3 V2.0

What's new ?

• Upload up to 4 images to the LLM for prompt generator and modificator

• "Audio Generator" replace the previous "Text Voice Dialogue Generator" (it allows now to generate an audio file from an user input or a prompt, great to control sound and voices when using ref2video)

The "Audio Generator" need your Minimax diffusion model (you don't need to download another model just use the one you usually use). Same for CLIP and Audio VAE.

Audio Generation uses LLM to generate the prompt then Minimax model at min resolution and then extract audio to a file. It's quick but depends on the hardware (avg 15-20 sec for 15 sec audio with RTX 5080).

Hope it helps some of you, let me know in the comments if you have some feedbacks !

Minimax H3 V1.0

This version has 4 tools:

• Prompt generator using image (if you use vision, you need to precise it in user input if the image is first frame or just a reference)

• Prompt Modificator using image

• Chatbot (if you need to chat with the LLM)

• Text Voice Dialogue Generator (usefull if you create your own audio dialogues using prompts)

The green cards in each groups explain how to use them.

Krea2 V1.0

Hello,

If like me you are fed up to switch between Comfyui and ollama/lmStudio, here is an easy workflow to use and config for 2 distincts needs:

• Prompt enhancer using an image as reference

• Prompt modificator to apply changes in an existing prompt (can use an image here too)

It's my first post so hope it will help someone !

PS: You can also use it with other AI than krea2, just edit both instructions prompts for your own rules.

I'm working on Krea2 V2.0, I will publish it when I can.