Instructions to use mihai-chindris/lora-workflow-template with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use mihai-chindris/lora-workflow-template with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("fill-in-base-model", dtype=torch.bfloat16, device_map="cuda") pipe.load_lora_weights("mihai-chindris/lora-workflow-template") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
Publishing Plan (code-only)
GitHub (Mihai Codes org)
Recommended: publish this project as a workflow repo without personal data.
- Use
README-public.mdas repository README. - Keep
.gitignoreas-is. - Verify no files under data/output/checkpoints are tracked.
- Push only workflow scripts and docs.
Hugging Face
Recommended: do not publish personal-face LoRA weights publicly.
Safer alternatives:
- Publish a Space or repo with training workflow docs only.
- Publish a template model card with no weights.
If you keep an existing personal model on HF:
- Prefer switching visibility to private first.
- Keep or delete based on your risk tolerance; if uncertain, keep private.
Existing HF model decision
For mihai-chindris/image-generator:
- If you do not actively need public access, set it to private now.
- Delete only if you are sure you never need it again.
Reason: it is identity-linked and publicly downloadable; private mode gives you immediate risk reduction without irreversible loss.