Synthyra / ESMFold2-Experimental-Cutoff2025

FastPLMs ESMFold2

FastPLMs ESMFold2 is a self-contained Hugging Face AutoModel wrapper for Biohub's ESMFold2 and ESMFold2-Fast structure predictors. It vendors the released Biohub ESMFold2 model code, ESMC backbone code, input builder, MSA helpers, and structure export utilities needed for remote-code loading.

Load With AutoModel

import torch
from transformers import AutoModel

model = AutoModel.from_pretrained(
    "Synthyra/ESMFold2-Fast",
    trust_remote_code=True,
    dtype=torch.float32,
).eval().cuda()

Use Synthyra/ESMFold2 for the full model and Synthyra/ESMFold2-Fast for the faster release variant. The folding trunk runs in fp32; the 6B ESMC backbone is loaded in bf16 by default via esmc_precision="bf16" .

Fold One Protein

sequence = "MKTLLILAVVAAALA"

result = model.fold_protein(
    sequence,
    num_loops=3,
    num_sampling_steps=50,
    num_diffusion_samples=1,
    seed=0,
)

print(float(result.plddt.mean()))
print(float(result.ptm))

Save mmCIF or PDB

model.save_as_cif(result, "prediction.cif")
model.save_as_pdb(result, "prediction.pdb")

cif_text = model.result_to_cif(result)
pdb_text = model.result_to_pdb(result)

result_to_cif preserves the full MolecularComplex . result_to_pdb converts through Biohub's protein-only ProteinComplex representation, so use mmCIF for complexes with ligands or nucleic acids.

Fold Complexes

types = model.input_types

complex_input = types.StructurePredictionInput(
    sequences=[
        types.ProteinInput(id="A", sequence="MKTLLILAVVAAALA"),
        types.DNAInput(id="B", sequence="GATAGC"),
        types.LigandInput(id="L", ccd=["SAH"]),
    ]
)

result = model.fold(
    complex_input,
    num_loops=3,
    num_sampling_steps=50,
    num_diffusion_samples=1,
    seed=0,
)

model.save_as_cif(result, "complex_prediction.cif")

Use MSAs

types = model.input_types

msa = types.MSA.from_a3m("query.a3m", max_sequences=128)
input_with_msa = types.StructurePredictionInput(
    sequences=[
        types.ProteinInput(id="A", sequence=msa.query, msa=msa),
    ]
)

result = model.fold(input_with_msa, num_sampling_steps=50, seed=0)

Raw Tensor Inference

features, chain_infos = model.prepare_structure_input(complex_input, seed=0)

with torch.inference_mode():
    output = model(
        **features,
        num_loops=3,
        num_sampling_steps=50,
        num_diffusion_samples=1,
    )

decoded = model.input_builder.decode(output, features, chain_infos)

Set load_esmc=False when loading if you want to provide precomputed lm_hidden_states manually or run folding-trunk tests without loading the 6B ESMC backbone:

model = AutoModel.from_pretrained(
    "Synthyra/ESMFold2-Fast",
    trust_remote_code=True,
    load_esmc=False,
).cuda().eval()