ProteinsGPU5 credits

ESMFold — Protein Structure Prediction

Predict 3D protein structure from sequence in seconds

ESMFold uses the ESM-2 protein language model to predict 3D protein structure directly from amino acid sequence. Unlike AlphaFold, it doesn't require multiple sequence alignments (MSA), making it 60x faster while maintaining high accuracy for single-chain predictions.

$0.05

per API call

credits per call

/v1/proteins/fold

API endpoint

Features

Predict 3D structure from amino acid sequence

Returns PDB file with atomic coordinates

Per-residue pLDDT confidence scores

No MSA required — single sequence input

GPU-accelerated on NVIDIA A100

Async job submission with polling

Quick Start

ESMFold — Python Examplepython

import requests

API_KEY = "sk-sci-your-key-here"
url = "https://scirouter.ai/v1/proteins/fold"

# Submit a folding job
response = requests.post(url, json={
    "sequence": "MKFLILLFNILCLFPVLAADNHGVS..."
}, headers={"Authorization": f"Bearer {API_KEY}"})

job = response.json()
print(f"Job ID: {job['job_id']}")

# Poll for results
import time
while True:
    result = requests.get(f"{url}/{job['job_id']}",
        headers={"Authorization": f"Bearer {API_KEY}"})
    data = result.json()
    if data["status"] == "completed":
        print(f"PDB file: {len(data['pdb'])} bytes")
        print(f"Mean pLDDT: {data['mean_plddt']:.1f}")
        break
    time.sleep(2)