Instructions to use Abiray/MiniMax-H3-Turbo-Lora-Pruned-ComfyUI with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Abiray/MiniMax-H3-Turbo-Lora-Pruned-ComfyUI with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Abiray/MiniMax-H3-Turbo-Lora-Pruned-ComfyUI", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
Very Sharp Results ??
seems that the video results with turbo, is very sharp!?
human skins is so feels plastic ...
so, is it possible to refine?
It really depends on the image you used or the prompt you wrote. And if it does end up looking plastic, just use a LoRA to fix it.
what LoRA?
and i mention that, your video sample is very sharp too :]
Yeah, I know. I used an AI-generated picture as input, and the video resolution was only 0.4MP without any LoRA other than turbo lora. We can improve the results by using a high-quality realistic reference image and prompt, along with a good LoRA from Civitai like the GalaxyAce LoRA.
dear author, would you kindly help to update your pruned lora for "minimax_h3_turbo_v4_step600_ema.safetensors" which is the most recommended version at this moment?
Sure