lore_generator.py

import torch
from transformers import pipeline
from utils import load_config

def generate_lore(seed_prompt):
    cfg = load_config()
    model_name = cfg['generation']['model_name']
    device = 0 if torch.cuda.is_available() else -1
    pipe = pipeline('text2text-generation', model=model_name, device=device)
    prompt = f"Provide a deep history and mythology for this world seed:\n{seed_prompt}\n\nRespond in markdown with timeline and myth summaries."
    out = pipe(prompt, max_new_tokens=400, do_sample=True, temperature=cfg['generation']['temperature'])
    return out[0].get('generated_text') if isinstance(out, list) else str(out)