Downloads · 30 days
0
LordXmen2k/Tulu-Geo-DiT-v1
Tulu-Geo-DiT-v1 is a text generation model from LordXmen2k. Use it when you need the model to write or continue text. It is set up for pytorch. The card lists the license as apache-2.0.
Tulu-Geo-DiT-v1 is a specialized Diffusion Transformer (DiT) architecture optimized for factual geography retrieval.
Downloads · 30 days
0
Access
Public
Updated Jan 2, 2026
Repo size
557 MB
Likes
0
Public
Click a slice to open those files.
.pt557 MB · 100%
From the Hugging Face model README
Tulu-Geo-DiT-v1 is a specialized Diffusion Transformer (DiT) architecture optimized for factual geography retrieval.
The model weights (tulu_geo_expert.pt) are designed to be used with the pipeline.py script.
from pipeline import generate_expert
# 150 Denoising Steps -> Factual Output
print(generate_expert("The capital of France is"))
If you use this model or code in your research, please cite it as follows:
@software{lordxmen2k_tulu_geo_2025,
author = {Gerald Enrique Nelson Mc Kenzie},
title = {Tulu-Geo-Diffusion: Factual Retrieval via Latent Denoising in Diffusion Transformers},
year = {2025},
publisher = {Zenodo},
version = {v1.0.0},
doi = {10.5281/zenodo.18030750},
url = {https://doi.org/10.5281/zenodo.18030750}
}