Skip to main content

gensyn-audit

gensyn-audit independently replays Gensyn Open 1B training steps and verifies the results against the run's published commitments.

Run an audit

The current Open 1B audit requires Python 3.11 and an Apple Silicon Mac.

Install the CLI:

uv tool install --python 3.11 gensyn-audit
gensyn-audit --version

Then go to:

https://open1b.gensyn.ai/

The site explains the available audits and their resource requirements, lets you claim one, and gives you the exact command to run. Do not share its claim token.

Run gensyn-audit --help for general CLI help.

What it verifies

Depending on the selected audit, gensyn-audit checks the downloaded training code and kernels, checkpoint integrity, and the replayed state against the run's published commitments. Claimed results are also compared with loss values withheld by the record.

A match means your machine independently reproduced the state hash Gensyn published for that training step. Each successful audit verifies one step of Open 1B's training history and contributes toward independently auditing the full run.

License

The audit client is open sourced under Apache 2.0.

Running an audit also uses RepOp, Gensyn's reproducibility library, which is separately governed by the Gensyn Reproducibility License v1.0. That license allows you to use RepOp to audit Open-1B's training and publish your findings.

Metadata

Release files for gensyn-audit 1.1.0

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

Source distribution (sdist)

Source distribution for gensyn-audit 1.1.0
File Size Uploaded Provenance Actions
gensyn_audit-1.1.0.tar.gz 207.4 kB Without provenance Details for gensyn_audit-1.1.0.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for gensyn-audit 1.1.0
File Interpreter ABI Platform Actions
gensyn_audit-1.1.0-py3-none-any.whl Python 3 none any Details for gensyn_audit-1.1.0-py3-none-any.whl

Total release size: 339.9 kB

Release files / gensyn_audit-1.1.0.tar.gz

Download URL gensyn_audit-1.1.0.tar.gz
Size 207.4 kB
Tags Source
SHA-256 checksum
How to use checksums
3e6b7195221fc3f7b6d572f2fbbc88f860fc807e4b55621854d939bed790f66e
BLAKE2b-256 checksum
How to use checksums
76d580f075cc9ab67c95530adedcbe346973b593955b890c22f9542ac1f879ab
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via uv/0.10.7 {"installer":{"name":"uv","version":"0.10.7","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

Release files / gensyn_audit-1.1.0-py3-none-any.whl

Download URL gensyn_audit-1.1.0-py3-none-any.whl
Size 132.5 kB
Tags Python 3
SHA-256 checksum
How to use checksums
4a08d3d35a4ab811ee014179094c4380cba0213210336e851061fce849ce51a0
BLAKE2b-256 checksum
How to use checksums
5693ed15ede8e1da6e8863e02ee836d4e88fa2bd10e43b7f045ff5f0b5c46f88
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via uv/0.10.7 {"installer":{"name":"uv","version":"0.10.7","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

Security for gensyn-audit 1.1.0

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.

No provenance found

None of the files in this release were uploaded with provenance metadata.

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.

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