Instructions to use c-bone/CrystaLLM-pi_Mattergen-XRD with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use c-bone/CrystaLLM-pi_Mattergen-XRD with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="c-bone/CrystaLLM-pi_Mattergen-XRD")

# Load model directly
from transformers import AutoTokenizer, SliderGPT

tokenizer = AutoTokenizer.from_pretrained("c-bone/CrystaLLM-pi_Mattergen-XRD")
model = SliderGPT.from_pretrained("c-bone/CrystaLLM-pi_Mattergen-XRD", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use c-bone/CrystaLLM-pi_Mattergen-XRD with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "c-bone/CrystaLLM-pi_Mattergen-XRD"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "c-bone/CrystaLLM-pi_Mattergen-XRD",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/c-bone/CrystaLLM-pi_Mattergen-XRD

SGLang

How to use c-bone/CrystaLLM-pi_Mattergen-XRD with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "c-bone/CrystaLLM-pi_Mattergen-XRD" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "c-bone/CrystaLLM-pi_Mattergen-XRD",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "c-bone/CrystaLLM-pi_Mattergen-XRD" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "c-bone/CrystaLLM-pi_Mattergen-XRD",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use c-bone/CrystaLLM-pi_Mattergen-XRD with Docker Model Runner:
```
docker model run hf.co/c-bone/CrystaLLM-pi_Mattergen-XRD
```

Model Card for CrystaLLM-pi_Mattergen-XRD

Model Details

Model Description

CrystaLLM-pi_Mattergen-XRD is a conditional generative model designed for the recovery of crystal structures from X-ray Diffraction (XRD) data. It is a fine-tuned version of the CrystaLLM-pi framework, based on a GPT-2 decoder-only architecture. This variant employs the Residual Attention (Slider) mechanism to condition the generation of Crystallographic Information Files (CIFs) on high-dimensional experimental data.

The model generates crystal structures based on an XRD pattern input vector, consisting of the 20 most intense peaks:

Peak Positions ($2\theta$)
Peak Intensities

Developed by: Bone et al. (University College London)
Model type: Autoregressive Transformer with Residual Attention Conditioning
Language(s): CIF (Crystallographic Information File) syntax
License: MIT
Finetuned from model: c-bone/CrystaLLM-pi_base

Model Sources

Repository: GitHub: CrystaLLM-pi
Paper: Discovery and recovery of crystalline materials with property-conditioned transformers (arXiv:2511.21299)
Dataset: HuggingFace: c-bone/mattergen_XRD

Uses

Direct Use

The model is intended for structure solution and recovery from powder XRD data. Researchers can input a list of peak positions and intensities derived from experimental diffraction patterns to generate candidate crystal structures that match the experimental signature.

Out-of-Scope Use

Disordered Systems: The model was trained on the alex-mp-20 dataset and theoretical XRDs. It does not natively handle partial occupancies or disorder.
Large Unit Cells: Context window limits apply (~20 atoms/cell).
Organic/MOFs: The training data only contains ordered organic crystals.

Bias, Risks, and Limitations

Missing Data: The "Slider" mechanism is designed to handle missing peaks (padded with -100), but significant data loss will degrade recovery rates.
Polymorphs: In cases of strong structural similarity or ambiguous diffraction patterns, the model may be biased towards the polymorph most represented in the training distribution.

Getting started

Generation from a scan: T2_load_and_generate.ipynb.

Citation

@misc{bone2025discoveryrecoverycrystallinematerials,
      title={Discovery and recovery of crystalline materials with property-conditioned transformers},
      author={Cyprien Bone and Matthew Walker and Bradley A. A. Martin and Kuangdai Leng and Luis M. Antunes and Ricardo Grau-Crespo and Amil Aligayev and Javier Dominguez and Keith T. Butler},
      year={2025},
      eprint={2511.21299},
      archivePrefix={arXiv},
      primaryClass={cond-mat.mtrl-sci},
      url={https://arxiv.org/abs/2511.21299},
}