Install

Requires Python 3.8+.

pip install fastfold-ai

Authentication

Set your API key in the environment:

export FASTFOLD_API_KEY="sk-...your-api-key"

Then create the client:

from fastfold import Client

client = Client()

You can also pass credentials explicitly:

from fastfold import Client

client = Client(api_key="sk-...your-api-key")

Base URL

By default, the SDK targets the Fastfold API. Override it when using a proxy, self-hosted deployment, or test environment:

from fastfold import Client

client = Client(base_url="https://api.fastfold.ai")

Services

The current SDK exposes:

• client.fold for the simplest fold-job flow
• client.jobs for raw payload creation, YAML submit, results, polling, visibility, and render helpers
• client.workflows for generic workflow creation, status, task results, execution, Evolla linked-history helpers, and workflow YAML endpoints
• client.library for item creation and file uploads
• client.openmm, client.openmmdl, client.evolla, and client.boltzgen for common multi-step workflow flows
• client.reports for Slack markdown reports

The package also installs the fastfold-cli entrypoint for shell-based usage:

fastfold-cli --help
fastfold-cli jobs --help
fastfold-cli workflows --help

Your first fold

from fastfold import Client

client = Client()

job = client.fold.create(
    sequence="LLGDFFRKSKEKIGKEFKRIVQRIKDFLRNLVPRTES",
    model="boltz-2",
    is_public=True,
)

print("Job ID:", job.id)

results = client.jobs.wait_for_completion(job.id, poll_interval=5.0, timeout=900.0)
print("Status:", results.job.status)
print("CIF URL:", results.cif_url())
print("Mean PLDDT:", results.metrics().mean_PLDDT)
print("Viewer:", results.get_viewer_link())

Next steps

• Use Overview for the SDK surface and recommended reading order
• Use CLI for shell-oriented commands
• Use Fold, OpenMM, OpenMMDL, Evolla, or BoltzGen for task-specific recipes