Text Generation
Transformers
Safetensors
English
gpt2
materials-science
crystallography
generative-ai
inverse-design
chemistry
xrd
text-generation-inference
Instructions to use c-bone/CrystaLLM-cXRD_mp20 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use c-bone/CrystaLLM-cXRD_mp20 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="c-bone/CrystaLLM-cXRD_mp20")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, PrefixXRDGPT tokenizer = AutoTokenizer.from_pretrained("c-bone/CrystaLLM-cXRD_mp20") model = PrefixXRDGPT.from_pretrained("c-bone/CrystaLLM-cXRD_mp20", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use c-bone/CrystaLLM-cXRD_mp20 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "c-bone/CrystaLLM-cXRD_mp20" # 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-cXRD_mp20", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/c-bone/CrystaLLM-cXRD_mp20
- SGLang
How to use c-bone/CrystaLLM-cXRD_mp20 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-cXRD_mp20" \ --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-cXRD_mp20", "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-cXRD_mp20" \ --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-cXRD_mp20", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use c-bone/CrystaLLM-cXRD_mp20 with Docker Model Runner:
docker model run hf.co/c-bone/CrystaLLM-cXRD_mp20
|
Download README.md from c-bone/CrystaLLM-cXRD_mp20: direct link, hf CLI and curl.
- Browser
- Download file 4.64 kB
-
https://huggingface.co/c-bone/CrystaLLM-cXRD_mp20/resolve/main/README.md
- Command line
-
hf download hf://c-bone/CrystaLLM-cXRD_mp20/README.md
-
curl -L -H "Authorization: Bearer $HF_TOKEN" -o README.md https://huggingface.co/c-bone/CrystaLLM-cXRD_mp20/resolve/main/README.md
4.64 kB
| language: | |
| - en | |
| license: mit | |
| library_name: transformers | |
| tags: | |
| - materials-science | |
| - crystallography | |
| - generative-ai | |
| - inverse-design | |
| - chemistry | |
| - xrd | |
| datasets: | |
| - c-bone/cxrd_mp20 | |
| pipeline_tag: text-generation | |
| # Model Card for CrystaLLM-cXRD\_mp20 | |
| ## Model Details | |
| ### Model Description | |
| **CrystaLLM-cXRD\_mp20** is a conditional generative model for crystal structure generation from powder X-ray Diffraction (XRD) data. It is a GPT-2 decoder-only model that conditions the generation of Crystallographic Information Files (CIFs) on a **continuous diffraction profile** rather than a peak list, using a **Perceiver Resampler prefix** (`PrefixXRD`) that compresses the pattern into learned key-value tokens injected at every transformer layer. | |
| The model consumes the **whole pattern**: | |
| 1. **Scattering vector** (Q) on a fixed 1000-point grid, `arange(0, 10, 0.01)` Å⁻¹ | |
| 2. **Intensity** (I), max-normalised to $[0, 1]$ | |
| This removes the peak-picking step from the workflow: a raw diffractometer scan is converted to the `(1000, 2)` `[Q, I]` profile automatically at generation time. Because Q is wavelength-independent, patterns from any anode can be used as long as the wavelength is supplied. | |
| Trained from scratch on MP-20, this is the **benchmark** model behind the MP-20 XRD-guided structure prediction comparison. For application use, [`c-bone/CrystaLLM-cXRD_chili100k`](https://huggingface.co/c-bone/CrystaLLM-cXRD_chili100k) (experimental nanomaterials) or [`c-bone/CrystaLLM-cXRD_alex-mp-20`](https://huggingface.co/c-bone/CrystaLLM-cXRD_alex-mp-20) (broad computed distribution) are the better choices. | |
| - **Developed by:** Bone et al. (University College London) | |
| - **Model type:** Autoregressive Transformer with Perceiver prefix conditioning (~36.5M parameters) | |
| - **Language(s):** CIF (Crystallographic Information File) syntax | |
| - **License:** MIT | |
| - **Finetuned from model:** none, trained from random initialisation | |
| ### Model Sources | |
| - **Repository:** [GitHub: CrystaLLM-pi](https://github.com/C-Bone-UCL/CrystaLLM-pi) (generation), [GitHub: CrystaLLM-cXRD](https://github.com/C-Bone-UCL/CrystaLLM-cXRD) (training and benchmarks) | |
| - **Paper:** To add | |
| - **Dataset:** [HuggingFace: c-bone/cxrd\_mp20](https://huggingface.co/datasets/c-bone/cxrd_mp20) | |
| ## Uses | |
| ### Direct Use | |
| Structure generation from powder XRD on the MP-20 distribution, and the reference point for the MP-20 benchmark. Pass a raw two-column scan and a target composition; the model proposes candidate structures consistent with the diffraction signature. | |
| ### Out-of-Scope Use | |
| - **Disordered Systems:** No native handling of partial occupancies or significant disorder. | |
| - **Experimental nanomaterials:** MP-20 is a computed, fully ordered distribution. Use the CHILI-100K model for experimental patterns. | |
| - **Unconditional generation:** The model requires a diffraction profile at every forward pass and has no missing-conditioning fallback. | |
| - **Non-CuKα data without a declared wavelength:** CuKα1 is assumed when none is given, which silently misplaces every peak in Q. | |
| ## Bias, Risks, and Limitations | |
| - **Computed-data bias:** Training patterns are simulated from relaxed DFT structures, so peak shapes are cleaner and more ordered than laboratory scans. | |
| - **Preprocessing sensitivity:** Recovery depends on the background subtraction and wavelength used to build the profile. Inspect the conversion (`--save_plot`) before trusting a result. | |
| - **Measured range:** Regions of the Q grid outside the measured scan are zero-filled; short scans give the model less to work with. | |
| - **Context window:** 1024 tokens (plus 8 prefix tokens), so very large unit cells will not fit. | |
| ## Getting started | |
| For generation from a scan, use [`T4_XRD_continuous.ipynb`](https://github.com/C-Bone-UCL/CrystaLLM-pi/blob/main/notebooks/T4_XRD_continuous.ipynb) in CrystaLLM-pi. For the benchmark this model was trained on, see [`B1_mp20_benchmark.ipynb`](https://github.com/C-Bone-UCL/CrystaLLM-cXRD/blob/main/notebooks/B1_mp20_benchmark.ipynb). | |
| ## Citation | |
| **Papers to cite:** | |
| ```bibtex | |
| @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}, | |
| } | |
| ``` | |
| and | |
| To add | |