Skip to main content

license: apache-2.0 library_name: numpy tags: [governed-ai, khipu, szl-holdings, receipts, lambda-gate]

SZL Holdings — Orbit v2

Orbit artwork: SZL Holdings / Stephen Lutar · CC BY 4.0 · Trademark notice. Software and artifact rights remain declared in the original metadata and documentation.

SZL Khipu

Inspect advisory NumPy kernels, small reference fixtures, and hash-chained run receipts.

Artifact: Python / NumPy software package · Stage: Software / reference

Explore in Command Lab · Build · Evidence

Before you use it

  • Λ is advisory: Conjecture 1 OPEN and proven_trust=false.
  • CUDA and energy measurement are UNAVAILABLE; software source binding does not establish a production deployment.
  • The package and Nano fixtures do not qualify a 1.5B checkpoint. Verify each artifact and its separate inference contract.
Technical details and evidence

szl-khipu

org: szl-holdings doctrine

Control before action. Evidence after.

Part of the szl-holdings estate · Product: a-11-oy.com · Proof: a11oy.net

Knot the run. Hash the proof. Fail closed.

Governed-AI khipu: an advisory Λ-gate, original YARQA canal attention, a TinyKhipu NAVIGATE/ABSTAIN silhouette, and hash-chained receipts. CPU numpy is LIVE. CUDA is UNAVAILABLE. energy UNAVAILABLE. proven_trust is always false.

Canonical GitHub source: szl-holdings/szl-khipu
Hub card: SZLHOLDINGS/szl-khipu
Kernels: SZLHOLDINGS/szl-khipu-kernels

The canonical publish-hf workflow mirrors szl_khipu/, README.md, LICENSE, and pyproject.toml from immutable Git blobs in one Hub model commit. Its MODEL_SOURCE_BINDING.json records the Git commit, tree, and each file digest; publication succeeds only after immutable provider readback matches every byte. Historical Hub files outside this software scope are preserved and are not attested by this binding. Software parity makes no trained-checkpoint claim.

Flagship55 Khipu research design — NOT A TRAINED MODEL

Three-specialist learning fabric · metadata-only manifest · fail-closed validator · offline triad packet · 91 public dataset references. The proposed student has YACHAY teacher, WILLAY guide and KHIPU skeptic adapter roles; none is trained, loaded, or deployed here. The manifest contains 13 review-only dataset examples; the separate index contains 91 source-pinned, public-visibility dataset handles from a dated Hub census. The 45 private identities remain outside this repository. Neither the index nor the packet grants dataset rights, model-training authority, independent evaluation, publication or production qualification.

What it is

  • A Python package (szl_khipu) that knots a run, hashes a receipt, and fail-closes on a zero axis or a broken chain.
  • Λ: weighted geometric mean over 13 Yuyay axes, advisory, never proven trust.
  • YARQA: contiguous canals; attend only inside the compartment; leak is the bound.
  • TinyKhipu-Nano: NAVIGATE / ABSTAIN silhouette. A few thousand floats. Hard ID filter.
  • ReceiptAgent-Nano: the reviewed szl_khipu/train/receipt_agent.py defines 0=ALLOW, 1=WARN, 2=BLOCKED, 3=ESCALATE. The deterministic rule_check decision remains authoritative; learned predictions do not override it. This software binding does not attest the sibling archive, its card, or its historical training/class-index contract.
  • Receipts: sha256 of weights, seed, steps, loss. Honesty REPORTED. Joules null.
  • Organ integrity: five-organ fail-closed kernel of szl-holdings/anatomy. HEART/YUYAY, YAWAR, YACHAY, OTel, Khipu skeleton. Not a Three.js rehost. The 3D atlas is SLSA L1 static viz.

What it is NOT

  • Not Qwen. Not a Qwen fine-tune, not a Qwen rehost, not a Qwen eval.
  • Not 1.5B. Nothing in this tree trains or publishes SZL-Khipu-1.5B, KHIPU-R2, or ReceiptAgent 1.5B. Those Hub cards stay RESEARCH and are not produced here.
  • Not a FlashAttention rehost. Not Dao .cu, not hopper/, not cute/, not Sage csrc, not vLLM paged .cu, not flex_attention.py.
  • Λ uniqueness is OPEN. Conjecture 1 is not a theorem. Unconditional uniqueness under kernel A1–A5 is machine-checked FALSE.
  • Energy UNAVAILABLE. Never a fabricated joule. MEASURED-NVML or honest null.
  • Not trained weights at 1–2B. Not GGUF. Not a CUDA speedup. Not proven trust.

Install

pip install -e .
pip install -e ".[gradio]"   # optional — Gradio 6 demo; gradio>=6.27.0,<7
pip install -e ".[torch]"    # optional — torch path; CUDA still UNAVAILABLE here

Requires Python ≥ 3.11. Runtime dep: numpy>=1.26. Apache-2.0.

Verify

pip install -e ".[test]"
python -m pytest -q tests                  # package suite (CI also runs it as `python -m unittest discover -s tests`)
szl-khipu verify artifacts/TRAINING_RECEIPT.json   # recompute the committed archives' sha256 (exit 1 on any mismatch)
(cd hardware/khipu-x1 && pip install -e ".[dev]" && python -m pytest -q)   # KHIPU-X1 software reference

The Gradio demo has its own pinned install and real-SDK smoke test; see spaces/README.md.

Python frontend + backend

Stdlib HTTP. Same process serves the holographic UI and runs the kernels. No Gradio required.

szl-khipu serve --host 0.0.0.0 --port 7860
# GET          /  /healthz  /version  /api/version  /.well-known/szl-source.json
# GET or POST  /api/lambda  /api/anatomy  /api/greenlight  /api/yarqa  /api/tiledigest
#              /api/prefix  /api/route  /api/bench  /api/infer

A malformed request is an explicit 4xx with its reason, never a default-filled 200. /api/infer reports which weights answered: the committed artifacts/*.npz (with its sha256) in a checkout or the Space, or a small on-request build when a wheel install has no artifacts/.

Gold = OPEN. Proof teal = LIVE. Never green-as-proven. Energy UNAVAILABLE. Conjecture 1 stays OPEN.

Static JS silhouette (not the NumPy kernel): docs/index.html · live at holdings.a-11-oy.com/khipu/ (served by szl-holdings/szl-holdings.github.io). The former khipu-pages repository is archived.

Quickstart

Lambda (advisory)

from szl_khipu import evaluate_lambda, YUYAY_FLOORS

ev = evaluate_lambda(list(YUYAY_FLOORS))
print(ev.value, ev.blocked, ev.reason)
# advisory pass — uniqueness remains Conjecture 1 OPEN

zeroed = list(YUYAY_FLOORS)
zeroed[0] = 0.0
print(evaluate_lambda(zeroed).blocked)  # True — fail closed

CLI:

szl-khipu demo-lambda
szl-khipu demo-lambda --zero 0

YARQA (canal leak)

import numpy as np
from szl_khipu import yarqa_attn

q = k = v = np.random.default_rng(7).standard_normal((12, 4))
out = yarqa_attn(q, k, v, n_canals=3)
print(out.leaked)  # bound: leaked ≤ 1e-9
szl-khipu demo-yarqa --n-canals 3

For numerical output without dense probability diagnostics, use yarqa_attn_output(q, k, v, n_canals=3). This NumPy path computes query tiles inside each canal and avoids full sequence-square score/probability matrices. The API contract and reproducible CPU comparison document validation, memory limits, attribution, and measurement boundaries at the reviewed source commit. This document is outside the model software mirror scope and is not included in that Hub model snapshot.

Organ integrity (five organs)

Fail-closed substrate. Any DOWN organ or a WILLAY veto blocks the body. Λ stays Conjecture 1 OPEN. Energy UNAVAILABLE. Locked-proven stays 8. Not a Three.js rehost.

from szl_khipu import evaluate_anatomy

ev = evaluate_anatomy(seed=11)
print(ev.live_count, ev.blocked, ev.energy)  # 5 False UNAVAILABLE

print(evaluate_anatomy(zero_heart=True).blocked)        # True
print(evaluate_anatomy(fabricate_joule=True).energy_j)  # None — never a fabricated joule
szl-khipu demo-anatomy
szl-khipu demo-anatomy --tamper-chain
szl-khipu demo-anatomy --fabricate-joule

Canonical anatomy repo: szl-holdings/anatomy · 3D atlas: SZLHOLDINGS/anatomy

Train TinyKhipu

szl-khipu train --seed 20260721 --steps 280 --out runs/tinykhipu
# writes runs/tinykhipu/tiny_khipu.npz + runs/tinykhipu/training_receipt.json
szl-khipu verify runs/tinykhipu/training_receipt.json
# exit 1 if a declared file is missing or its sha256 differs

--out defaults to runs/tinykhipu under the working directory. Keep runs out of artifacts/, which holds the committed, Hub-published export.

from szl_khipu import train_tiny_khipu

result = train_tiny_khipu(seed=20260721, steps=280)
print(result.plan_valid, result.abstain, result.hallucinated)

Export all three silhouettes (TinyKhipu, ReceiptAgent, moons), the MiniEmbed table and a combined receipt:

python scripts/train_and_export.py --out runs/export
# runs/export/*.npz + runs/export/TRAINING_RECEIPT.json
# honesty REPORTED · energy_j null · proven_trust false · whatNot listed
szl-khipu verify runs/export/TRAINING_RECEIPT.json

Without --out the script rewrites the committed artifacts/ in place. Those four archives are the historical 2026-08-29 export; the four Nano Hub repos and the Space carry the same bytes (their LFS SHA-256 values equal artifacts/TRAINING_RECEIPT.json). Current source does not reproduce those file digests: TinyKhipu's arrays changed with the curriculum fix in #103, and ReceiptAgent, moons and MiniEmbed regenerate identical arrays inside a differently versioned zip container, so their file SHA-256 differs too.

ReceiptAgent keeps the historical 600 training / 200 development examples and selects weights and stops using development agreement. Its agree and n_held return values are compatibility aliases for development_agree and n_development; they are not independent test results. After selection, the trainer measures test_agree once on 400 additional synthetic examples from a separate deterministic random stream. Duplicate feature rows within or across all three splits are rejected before training, independently of labels. New export receipts include both scores, split counts, and explicit metric roles. Historical receipts and weights are not rewritten by this source change.

This test measures the same synthetic generator, not unseen source families or production model quality. Repeated runs on the same seed reuse the same test examples, so that test must not become a hyperparameter-selection target. The deterministic rule_check remains authoritative, including when the surrogate disagrees; this evaluation grants no GPU-training or production admission.

Kernels vs field leaders (honest deltas)

SZL cut Field leader Honest delta
TileReceipt (szl-receipt-attn) FlashAttention (Dao et al., NeurIPS 2022) Numeric silhouette of tiled fused attention. SHA receipt of the tiles. No speedup claim. Not a rehost of Dao .cu / hopper / cute.
BlockWitness (szl-block-kv) PagedAttention / vLLM (Kwon et al., SOSP 2023) Paged-KV gather with a digest of the block table. No tokens/s. Not a vLLM rehost.
ScoreMod Fiber (szl-maskmod) FlexAttention (PyTorch, arXiv:2412.05496) Original score_mod + block-mask path that receipts the mask digest. No CUDA benches. A swapped mask cannot stay silent.
YARQA-ATTN SageAttention (Zhang et al., ICLR 2025) Not quantized attention. Contiguous canals (Quechua yarqa), attend inside the compartment, receipt the partition. GPU cubins UNAVAILABLE. CPU numpy path LIVE.
Governed RMSNorm RMSNorm (Zhang & Sennrich, 2019) / LayerNorm Matches a reference and optionally chains a digest of the rounded output. No speedup. Digest is integrity, not authorship.
Λ-gate UN HDI 2010 WGM / OECD composite indicators WGM over [0,1] axes, fail-closed on any zero or non-finite axis, A1–A4 runtime checks. Uniqueness is Conjecture 1 — OPEN. Advisory, never proven trust.
govsign + provctl in-toto / SLSA / Sigstore / DSSE Governance predicate (Λ, energy, decision) as an envelope. proven_trust locked false. UNSIGNED is a first-class fallback, never a fake signature. SLSA L1 honest, L2 roadmap.

This repo silhouettes Λ, YARQA, TinyKhipu, ReceiptAgent, and receipts in numpy. It does not replace those field leaders and does not inherit their benches.

Locked-8

Lean locked-proven formulas = exactly 8. Lab numerics CHECKED ≠ Lean PROVEN. The F-number mapping onto any larger registry is UNKNOWN — never fabricated.

ID Name Caveat
F1 Replay-Hash Determinism
F4 Khipu DAG acyclicity edges dst < src
F7 Chaski FIFO drain(enqueueAll([], msgs)) = msgs
F11 Ayni Reciprocity Conservation Σ in = Σ out
F12 Kuramoto boundedness additive fragment ONLY — not full nonlinear sync
F18 RS(10,6) recovery recoverable iff ≥ 6 of 10 shards
F19 Bekenstein additive scaffolding monotone fragment ONLY — not S ≤ 2πkRE/ℏc
F22 Khipu emit monotonicity sequence numbers strictly increase

Conjecture 1

Any two aggregators satisfying A1–A4 agree on every input. OPEN (sorry). Unconditional uniqueness under kernel A1–A5 is machine-checked FALSE.

Λ is advisory. Trust ceiling 0.97. proven_trust is false in this artifact and every receipt it emits.

Axioms carried at runtime (checks, not a uniqueness proof):

ID Name
A1 IsMonotone
A2 IsHomogeneous Λ(c·x)=c·Λ(x)
A3 IsEgyptianExact Λ(c,…,c)=c
A4 IsBounded Λ(x)≤max(x)
A5 IsPermutationInvariant

Honesty

Claim Label
CPU numpy path LIVE
CUDA / GPU cubins UNAVAILABLE
Energy / joules UNAVAILABLE (never fabricated)
Λ uniqueness Conjecture 1 OPEN
proven_trust false
TinyKhipu / ReceiptAgent / moons trained here REPORTED silhouette, not 1.5B
SZL-Khipu-1.5B / ReceiptAgent 1.5B RESEARCH, not trained in this tree
Weights at Hub 1–2B not applicable

Doctrine v11 LOCKED · 749 declarations · 14 axioms · 163 tracked sorries · locked-proven 8.

Hub siblings

Artifact What
hf/TinyKhipu-Nano NAVIGATE/ABSTAIN silhouette
hf/ReceiptAgent-Nano 4-way gate; kernel is truth
hf/Moons-Nano 2→8→2 two-moons fixture
hf/MiniEmbed-Nano 64 × 12 hash-token table, not a neural embedding
hf/szl-khipu-kernels Kernel Hub card; get_kernel
space/ Hub docker hologram — stdlib HTTP, no Gradio
spaces/ Gradio 6 demo (GitHub only; separate Docker publisher uses space/)

License

Apache-2.0. Copyright 2026 SZL Holdings.

Stephen P. Lutar Jr. / SZL Holdings · ORCID 0009-0001-0110-4173


Work with SZL Holdings

szl-khipu participates in the SZL governed-AI estate. The flagship control plane targets governed agent change management, but production/runtime qualification is tracked separately through the GitHub → Hugging Face → product → proof evidence chain. This repository does not itself prove or authorize a production deployment.

Metadata

Release files for szl-khipu 0.2.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for szl-khipu 0.2.0
File Size Uploaded Provenance Actions
szl_khipu-0.2.0.tar.gz 221.0 kB With provenance Details for szl_khipu-0.2.0.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for szl-khipu 0.2.0
File Interpreter ABI Platform Actions
szl_khipu-0.2.0-py3-none-any.whl Python 3 none any Details for szl_khipu-0.2.0-py3-none-any.whl

Total release size: 363.2 kB

Release files / szl_khipu-0.2.0.tar.gz

Download URL szl_khipu-0.2.0.tar.gz
Size 221.0 kB
Tags Source
SHA-256 checksum
How to use checksums
fa7ea63e1ac805f5b548effac3ce9c2415dba3925e26d0235d1c81d8fbd414cc
BLAKE2b-256 checksum
How to use checksums
e1e02c74fbaf81b9adb2a0fb392dfa109f28d226e62befe9b3577f4a2c18538b
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 18, 2026.

Transparency log

Release files / szl_khipu-0.2.0-py3-none-any.whl

Download URL szl_khipu-0.2.0-py3-none-any.whl
Size 142.2 kB
Tags Python 3
SHA-256 checksum
How to use checksums
df2ec92c02348cb9397f0203fbbf1e178e349f1121e1a1974e9ee16242fda3ea
BLAKE2b-256 checksum
How to use checksums
67e698e883945696d3af13c3890f6d30fba6ac6dfb2536d9da58d44a9a3bade2
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 18, 2026.

Transparency log

Security for szl-khipu 0.2.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Full provenance

Both files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for szl_khipu-0.2.0.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 18, 2026.

Transparency log

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

0.2.0 This release

2 release files Full provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page