Instructions to use Mubeiya/jarvis-janus-prompt-lora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use Mubeiya/jarvis-janus-prompt-lora with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("deepseek-ai/Janus-Pro-7B") model = PeftModel.from_pretrained(base_model, "Mubeiya/jarvis-janus-prompt-lora") - Notebooks
- Google Colab
- Kaggle
Jarvis Janus Prompt LoRA
This repository contains the trained Jarvis prompt-reverse LoRA adapter for deepseek-ai/Janus-Pro-7B.
It is used by Jarvis to reverse anime images into prompts with stronger artist-string handling and output formatting for Anima and NovelAI workflows.
The public Janus-Pro-7B base model is downloaded separately by Jarvis when missing. This repository only hosts the trained adapter and its sidecar tokenizer/config files.
- Downloads last month
- 8
Model tree for Mubeiya/jarvis-janus-prompt-lora
Base model
deepseek-ai/Janus-Pro-7B