THGLab/Llama-3.1-8B-GeomLlama-xyz

Llama-3.1-8B-GeomLlama-xyz

Built with Llama. Built with Axolotl.

GeomLlama-xyz is a fine-tune of Llama-3.1-8B-Instruct that generates 3D molecular conformer geometries directly from a SMILES string, emitting each structure as Cartesian XYZ coordinates (one element x y z line per atom). It is one of two models from our paper; the companion model, Llama-3.1-8B-GeomLlama-zmatrix, emits Fenske–Hall Z-matrix internal coordinates instead.

The model was trained jointly ("hybrid") on GEOM-QM9 and GEOM-Drugs, so it covers both small molecules and larger drug-like molecules with a single set of weights.

Quick start

The model was trained in the Alpaca instruction format. Reproduce the exact inference prompt used for the paper's numbers:

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "THGLab/Llama-3.1-8B-GeomLlama-xyz"
tok = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.bfloat16, device_map="auto")

smiles = "Cc1cccc(CSc2nnnn2-c2ccccc2)c1"
prompt = (
    "### Instruction:\n"
    "You can generate accurate molecular coordinates from a prompt "
    "containing a SMILES string.\n\n"
    "### Input:\n"
    "Generate a realistic equilibrium geometry for the molecule with the "
    f"following SMILES string in xyz format: {smiles}\n\n"
    "### Response:\n"
)

inputs = tok(prompt, return_tensors="pt").to(model.device)
out = model.generate(**inputs, max_new_tokens=3072, do_sample=True, temperature=1.0, top_p=0.95)
print(tok.decode(out[0][inputs.input_ids.shape[1]:], skip_special_tokens=True))

Sample many completions per SMILES (each is one candidate conformer) to build a conformer ensemble. T=1.0, top_p=0.95 and T=1.2, top_p=0.95 are good defaults.

Output format

One atom per line, element x y z in Ångström:

C -4.344237 -2.044144 -0.978303
C -3.501234  ...
...

Parse directly as an XYZ block (no header row is emitted).

View z-matrix or xyz coordinates easily at doublemolview.streamlit.app.

Training

• —Base: meta-llama/Llama-3.1-8B-Instruct
• —Framework: Axolotl
• —Method: LoRA (r = 32, α = 16, dropout 0.05, all linear layers), merged into the base weights
• —Epochs: 4 · LR: 3e-4, cosine · Optimizer: adamwbnb8bit
• —Sequence length: 4096, sample packing · Precision: bf16
• —Data: GEOM-QM9 + GEOM-Drugs, Cartesian XYZ targets (ori_xyz), plus the Alpaca instruction dataset for general-instruction rehearsal

Citation

Paper: How Well Can Frontier Large Language Models Generate Structures? High Quality Prediction of Molecular Geometries with Help from Fine-Tuning — arXiv:2607.13350. Please cite the paper and the underlying GEOM dataset (Axelrod & Gómez-Bombarelli, Scientific Data, 2022) if you use this model.

@misc{cavanagh2026geomllama,
  title         = {How Well Can Frontier Large Language Models Generate Structures?
                   High Quality Prediction of Molecular Geometries with Help from Fine-Tuning},
  author        = {Cavanagh, Joseph M. and Arnold, Jonathan B. and Alteri, Giovanni Battista
                   and Gritsevskiy, Andrew and Head-Gordon, Teresa},
  year          = {2026},
  eprint        = {2607.13350},
  archivePrefix = {arXiv},
  url           = {https://arxiv.org/abs/2607.13350}
}

License

Governed by the Llama 3.1 Community License. By using this model you agree to its terms. Built with Llama.