import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("TheRealPilot638/animagine-xl-4.0-inference", torch_dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]animagine-xl-4.0 — custom Inference Endpoint handler
Thin custom-handler repo (no weights) that serves
cagliostrolab/animagine-xl-4.0
with explicit width/height support — the stock SDXL toolkit container 400s on
non-square sizes. Used by the comfyui-orchestrator to render TEM portrait art
(scene/sprite 832×1216, tiger 896×1152, covers 896×1344).
Request: {"inputs": "<prompt>", "parameters": {negative_prompt, width, height, num_inference_steps, guidance_scale, seed}}
- Downloads last month
- -
Model tree for TheRealPilot638/animagine-xl-4.0-inference
Base model
stabilityai/stable-diffusion-xl-base-1.0 Finetuned
cagliostrolab/animagine-xl-4.0