Skip to content
ModelVerifiedExternal
Download

GLACIER 100k MiniMol

GLACIER is a multimodal student-teacher foundation model designed for molecular property prediction. It integrates molecular graphs, SMILES strings, and physicochemical descriptors to learn rich molecular embeddings.

Sourcehuggingface.co/glacier-hf/GLACIER-100k-MiniMol/tree/8bf9191e68c8b15f632a8bd4bd7210eba9f04066
KindOtherVersionRevision 8bf9191Publisher@glacier-hfCgrade

Model source

Kind
Other
File size
23.9 MB
Version
vRevision 8bf9191
Updated
Oct 7, 2026
Publisher
@glacier-hf
Downloads
2.6K
Source
Hugging Face

Source description

GLACIER: Graph-Language Alignment for Chemical Inference and Exploration using Representations

GLACIER is a multimodal student-teacher foundation model designed for molecular property prediction. It integrates molecular graphs, SMILES strings, and physicochemical descriptors to learn rich molecular embeddings.

Sample Usage

Since this model uses a custom architecture, you need to download the repository files to load the model.

import torch 
from huggingface_hub import snapshot_download
import sys

# Download the repository to access custom model code
repo_dir = snapshot_download(repo_id="glacier-hf/GLACIER-100k-MiniMol")
sys.path.append(repo_dir)

from data.dataloader import SmilesMoleculeDataset, build_dataloader
from glacier_student import Glacier

# Load the pretrained GLACIER model
model = Glacier.from_pretrained("glacier-hf/GLACIER-100k-MiniMol")

# Prepare input data
dataset = SmilesMoleculeDataset(smiles=["Cn1c(=O)c2c(ncn2C)n(C)c1=O"])
dataloader = build_dataloader(dataset, batch_size=1)

model.eval()
batch = next(iter(dataloader))
with torch.no_grad():
    embedding = model(batch)
print(embedding)

GLACIER Model Files

  • dataloader: customized dataloader for multimodal learning
  • encoders: graph, text, and tabular encoders
  • fusion: Finsler geometry-aware fusion method
  • glacier_student: GLACIER model backbone and contrastive loss
  • utils: miscellaneous helper functions

Citation

@inproceedings{nguyen2026glacier,
  title={GLACIER: A Multimodal Student-Teacher Foundation Model for Molecular Property Prediction},
  author={Emily Nguyen and Yongchan Hong and Harsh Toshniwal and Yan Liu and Andreas Luttens},
  booktitle={Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2 (KDD ’26)},
  year={2026},
  publisher={ACM},
  doi={10.1145/3770855.3819032}
}

Sources

1 source
Hugging FacePrimaryVersion 8bf9191e68c8b15f632a8bd4bd7210eba9f04066
huggingface.co/glacier-hf/GLACIER-100k-MiniMol/tree/8bf9191e68c8b15f632a8bd4bd7210eba9f04066

Verified Aug 30

Model artifacts

1 artifact

pytorch_model.bin

bin · 23.9 MB · SHA-256 4d6f516982a5…e124 · Hugging Face

Download

Source excerpts

3 excerpts
ReadmeJul 11huggingface.co/glacier-hf/GLACIER-100k-MiniMol

--- license: mit pipeline_tag: feature-extraction tags: - chemistry - multimodal --- # GLACIER: Graph-Language Alignment for Chemical Inference and Exploration using Representations GLACIER is a multimodal student-teacher foundation model designed for molecular property prediction. It integrates molecular graphs, SMILES strings, and physicochemical descriptors to learn rich molecular embeddings. - **Paper:** [GLACIER: A Multimodal Student-Teacher Foundation Model for Molecular Property Prediction](https://huggingface.co/papers/2606.11382) - **Repository:** [https://github.com/eemokey/glacier](https://github.com/eemokey/glacier) ## Sample Usage Since this model uses a custom architecture, you need to download the repository files to load the model. ```python import torch from huggingface_hub import snapshot_download import sys # Download the repository to access custom model code repo_dir = snapshot_download(repo_id="glacier-hf/GLACIER-100k-MiniMol") sys.path.append(repo_dir) from data.dataloader import SmilesMoleculeDataset, build_dataloader from glacier_student import Glacier # Load the pretrained GLACIER model model = Glacier.from_pretrained("glacier-hf/GLACIER-100k-MiniMol") # Prepare input data dataset = SmilesMoleculeDataset(smiles=["Cn1c(=O)c2c(ncn2C)n(C)c1=O"]) dataloader = build_dataloader(dataset, batch_size=1) model.eval() batch = next(iter(dataloader)) with torch.no_grad(): embedding = model(batch) print(embedding) ``` ## GLACIER Model Files - `dataloader`: customized dataloader for multimodal learning - `encoders`: graph, text, and tabular encoders - `fusion`: Finsler geometry-aware fusion method - `glacier_student`: GLACIER model backbone and contrastive loss - `utils`: miscellaneous helper functions ## Citation ```bibtex @inproceedings{nguyen2026glacier, title={GLACIER: A Multimodal Student-Teacher Foundation Model for Molecular Property Prediction}, author={Emily Nguyen and Yongchan Hong and Harsh Toshniwal and Yan Liu and Andreas Luttens}, booktitle={Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2 (KDD ’26)}, year={2026}, publisher={ACM}, doi={10.1145/3770855.3819032} } ```

NoteAug 30huggingface.co/glacier-hf/GLACIER-100k-MiniMol

Source context: 2585 downloads · 0 likes · Pipeline feature-extraction · Repo glacier-hf/GLACIER-100k-MiniMol

NoteJul 17huggingface.co/glacier-hf/GLACIER-100k-MiniMol

Source context: 394 downloads · 0 likes · Pipeline feature-extraction · Repo glacier-hf/GLACIER-100k-MiniMol