README.md

---
license: cc-by-nc-4.0
extra_gated_fields:
  First Name: text
  Last Name: text
  Date of birth: date_picker
  Country: country
  Affiliation: text
  I accept the terms and conditions: checkbox
  geo: ip_location
language:
- en
---

# Adjoint Sampling: Highly Scalable Diffusion Samplers via Adjoint Matching

This repository contains the models for our [official code release](https://github.com/facebookresearch/adjoint_sampling).  
Additionally, it contains the specific [eSEN](https://arxiv.org/abs/2502.12147) checkpoint that we used to train those models.  

**[Arxiv](https://arxiv.org/abs/2504.11713) (Accepted in ICML 2025)**  
*Authors: _Aaron Havens, Benjamin Kurt Miller, Bing Yan, Carles Domingo-Enrich, Anuroop Sriram, Brandon Wood, Daniel Levine, Bin Hu, Brandon Amos, Brian Karrer, Xiang Fu, Guan-Horng Liu, Ricky T. Q. Chen_*
![image/png](https://cdn-uploads.huggingface.co/production/uploads/67cd3633d2fd0e9ec4cbb329/8J1MzaVvtHjv4e5AY-S8B.png)

### Download models
```python
from huggingface_hub import snapshot_download

snapshot_download(
    repo_id="facebook/adjoint_sampling",
    local_dir="models",
    allow_patterns=[
        "am/",
        "bmam/",
        "pretrain_for_bmam/",
        "torsion/",
        "esen_spice.pt",
    ],
)
```