import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("BreakpointAI/socknet3", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]socknet3
Joint image + bounding-box diffusion model with a LoRA adapter.
Released by Breakpoint AI as part of open-sourcing the company's research artifacts.
| Training data | BreakpointAI/breakpoint-grounding-55m |
| Checkpoint step | 625,000 |
| Training run | W&B |
Contents
| Path | Size | What it is |
|---|---|---|
boxnet/ |
3.9 GB | Joint image + bounding-box diffusion backbone |
pytorch_lora_weights.safetensors |
305.2 MB | LoRA adapter weights |
Inference weights only. Optimizer, LR scheduler, RNG and dataloader state were not uploaded, so this checkpoint cannot be used to resume training.
Citation
@misc{breakpoint_socknet3,
title = {socknet3},
author = {Wang, Franklin and Zhong, Desmond and Murdoch, Jamie},
year = {2026},
url = {https://huggingface.co/BreakpointAI/socknet3}
}
- Downloads last month
- -