How to use from the
Use from the
Kernels library
# !pip install kernels

from kernels import get_kernel

kernel = get_kernel("SZLHOLDINGS/szl-provctl")

szl-provctl

Provenance-DAG control + in-toto / SLSA interop. Not a model. No weights.

IS: a software kernel that (1) emits declared in-toto Statement v1 and SLSA provenance v1 shapes from a UnifiedReceiptChain, (2) walks a provenance DAG edge-by-edge, (3) can bind a per-kernel NVML energy delta when a GPU is present. Public API: statement_from_chain, slsa_statement, ProvenanceDAG / verify_dag, measure_kernel_energy, selfcheck.

IS NOT: trained weights. Not a complete signing product (signing is szl-govsign). Not a CUDA speedup. A hash-chain digest is an integrity fingerprint, not a signature. External verifier compatibility must be tested. Hub model.joblib is QUARANTINED executable serialization — do not joblib.load it.

Canonical GitHub source: https://github.com/szl-holdings/szl-provctl
Hub package: https://huggingface.co/kernels/SZLHOLDINGS/szl-provctl
This model-type repo is the publish / card mirror. Apache-2.0.

from kernels import get_kernel
pc = get_kernel("SZLHOLDINGS/szl-provctl", revision="main", trust_remote_code=True)
print(pc.selfcheck()["ok"])
chain = pc.UnifiedReceiptChain()
chain.emit("governed_norm", "rms_norm", {"in_shape": [4, 64], "eps": 1e-6})
stmt = pc.statement_from_chain(chain, lambda_score=0.9, decision="ALLOWED")

Energy is MEASURED-only: no GPU/NVML → joules=None / UNAVAILABLE_NO_NVML, never fabricated. BLOCKED nodes stay in the DAG. proven_trust is structurally locked False.

Λ = Conjecture 1 (advisory, never a theorem). Doctrine v11. Copyright 2026 SZL Holdings · Stephen P. Lutar · ORCID 0009-0001-0110-4173.

Downloads last month
-
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Spaces using SZLHOLDINGS/szl-provctl 2

Collections including SZLHOLDINGS/szl-provctl