import spaces import torch from diffusers import DiffusionPipeline import gradio as gr # Load the pre-trained diffusion model pipe = DiffusionPipeline.from_pretrained('ptx0/terminus-xl-velocity-v2', torch_dtype=torch.bfloat16) pipe.to('cuda') # Define the image generation function with adjustable parameters and a progress bar @spaces.GPU def generate(prompt, guidance_scale, guidance_rescale, num_inference_steps, negative_prompt): return pipe( prompt, negative_prompt=negative_prompt, guidance_scale=guidance_scale, guidance_rescale=guidance_rescale, num_inference_steps=num_inference_steps ).images # Example prompts to demonstrate the model's capabilities example_prompts = [ ["A futuristic cityscape at night under a starry sky", 7.5, 25, "blurry, overexposed"], ["A serene landscape with a flowing river and autumn trees", 8.0, 20, "crowded, noisy"], ["An abstract painting of joy and energy in bright colors", 9.0, 30, "dark, dull"] ] # Create a Gradio interface iface = gr.Interface( fn=generate, inputs=[ gr.Text(label="Enter your prompt"), gr.Slider(1, 20, step=0.1, label="Guidance Scale", value=11.5), gr.Slider(0, 1, step=0.1, label="Rescale classifier-free guidance", value=0.7), gr.Slider(1, 50, step=1, label="Number of Inference Steps", value=25), gr.Text(value="underexposed, blurry, ugly, washed-out", label="Negative Prompt") ], outputs=gr.Gallery(height=1024, min_width=1024, columns=2), examples=example_prompts, title="Terminus XL Velocity v2.0 Demonstration", description="Terminus XL is a v-prediction model trained with a zero-terminal SNR noise schedule, allowing it to create very dark or very bright images. Occasionally, it will work pretty well for typography, eg. text in images." ).launch()