Using Vakint from Python

Symbolica 2.2.0 bundles Vakint’s community module. This matching-only workflow canonicalizes one loop with arbitrary input labels and invokes no external evaluation tool.

Install and verify the module

python -m venv .venv
. .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install "symbolica==2.2.0"
python -c "import symbolica.community.vakint as vakint; print(vakint.__name__)"

There is no separate vakint Python wheel. Follow Symbolica’s installation and license terms.

Canonicalize a one-loop integral

Save this as vakint_quickstart.py:

from symbolica import E
from symbolica.community.vakint import Vakint

engine = Vakint(evaluation_order=[])
integral = E(
    "topo(prop(18,edge(7,7),k(99),muvsq,1))",
    default_namespace="vakint",
)
canonical = engine.to_canonical(integral, short_form=True)

assert "I1L" in str(canonical)
print(canonical)

Run python vakint_quickstart.py. The arbitrary propagator, edge, and momentum labels are normalized to Vakint’s one-loop topology. An empty evaluation_order is intentional: it prevents this first use from probing FORM, MATAD, FMFT, or pySecDec.

Use the Rust guide for the native matching API, or continue with the matching and evaluation guide before enabling a backend.