dreamboothed stable diffusion 1.5 to include the concept of _alex_ person based on 4 photos shared in the Intel Insiders Discord.
published with permission.
Usage example
from diffusers import StableDiffusionPipeline
model_cp = "Vipitis/SD-15-dreamed-alex-person"
pipe = StableDiffusionPipeline.from_pretrained(model_cp)
out = pipe(prompt="photo of _alex_ person", negative_prompt="additional fingers, jepg artefacts", num_inference_steps=35, guidance_scale=8.5)
out.images[0]
Running on Intel Arc (via optimum-intel), has to convert and compile the model on first attempt.
from optimum.intel.openvino import OVStableDiffusionPipeline
model_cp = "Vipitis/SD-15-dreamed-alex-person"
ov_pipe = OVStableDiffusionPipeline.from_pretrained(model_cp, export=True, device="GPU")
out = ov_pipe(prompt="photo of _alex_ person", negative_prompt="additional fingers, jepg artefacts", num_inference_steps=35, guidance_scale=8.5)
out.images[0]
- Downloads last month
- 4
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.