Kolmogorov–Arnold Networks (KANs) are a recently proposed neural architecture in which every weightis replaced by a learnable univariate B-spline, grounded in the Kolmogorov–Arnold representationtheorem. Despite growing academic interest, the ecosystem has lacked a production-ready, multi-backendimplementation suitable for both rigorous experimentation and real-world deployment. We present KANX(kanx), the first fully-documented, cross-framework KAN toolkit: a pip-installable library (TensorFlowprimary, PyTorch secondary) with real ONNX export, a FastAPI REST service, Docker/Kubernetesmanifests, a 113-test suite (94% coverage), continuous integration, and automated PyPI releases. UsingKANX's reproducible benchmark harness, we conduct a rigorous, multi-baseline comparison of KANsagainst parameter-matched MLPs. On a smooth 2-D synthetic regression target—the best-case regime forKAN theory—a KAN [2,16,1] achieves a test MSE of 2.14 × 10⁻⁵ with only 432 parameters. We furtherreport five-fold cross-validated results on the UCI Diabetes dataset, with KAN-TF achieving R² = 0.449 ±0.130 with only 1,068 parameters—competitive with Ridge regression and substantially ahead of aparameter-matched MLP (R² = 0.089). All benchmarks were validated on a T4 GPU (Google Colab, June2026). We explicitly characterize the regime boundaries: KANs excel on smooth, low-dimensional,separable targets but incur 3–5× higher inference latency than equivalent MLPs. KANX closes the toolinggap, enabling fair comparisons and broader adoption of KAN research