Skip to main content

hf-simulation

Stochastic high-frequency seismogram generation: a Rust port of EMOD3D's hb_high_v6.0.3 (BINMOD + VERSION1), exposed to Python as a batched API.

import numpy as np
from hf_simulation import (
    FaultSegment,
    HfConfig,
    RecordParameters,
    SlipModel,
    Simulator,
    VelocityModel1D,
    station_seeds,
)

# The four groups mirror the simulation core's own decomposition, so a
# configuration written down elsewhere deserialises straight into this.
config = HfConfig(record=RecordParameters(duration_s=40.0))

# Built once per source: the air layer, the slip-model normalisation and the
# moment scaling do not depend on where the receiver is.
simulator = Simulator(SlipModel([segment]), velocity_model, config)

waveform = simulator.run_stations(  # (3, n_station, n_time), cm/s²
    latitude_deg=latitudes,
    longitude_deg=longitudes,
    station_seed=station_seeds(1234, station_names),
    kappa_s=kappas,  # one per station
    q_frequency_exponent=q_exponents,
)

Installation

pip install hf-simulation            # core: numpy only
pip install "hf-simulation[xarray]"  # adds hf_simulation.dataset.to_dataset

Status

Statistically equivalent to the production Fortran: intensity-measure means agree within ±2% at 373 of 375 endpoints tested (0 disagree; the remaining two are undetermined for want of sample size). Distribution shape agrees within ±4% on the 5th/50th/95th percentiles, and inter-frequency correlation passes Holm-corrected at 0 of 15. Each test is reported beside a null control — the same test with both sides drawn from the same program — so its false-alarm rate is measured rather than assumed. On the null, the shape test flags 11 of 375 at ±2% and none at ±4%, which is why the shape band is ±4%.

Output is not bit-identical to the Fortran: the FFT, the gamma function, the geodesic (WGS84), the pi constants and the RNG seeding all differ, and a few defects in the original are fixed. Each difference was checked statistically against the production output.

The batched interface

Simulator.run_stations takes arrays of station coordinates and seeds, plus the per-station site and path parameters (kappa_s, q_frequency_exponent), and returns every station at once. Everything else in HfConfig — including fmax_hz, the ray set and the path-duration model — is shared by the whole run. The stress parameter is per subfault: each FaultSegment takes a stress_drop_bars grid alongside slip. A station's seed is a uint64 derived by station_seeds through numpy.random.SeedSequence from a root seed and the station name, so stations are independent. Three properties follow, all of them tested:

  • station order changes no waveform — so dask rechunking is free;
  • simulating a subset equals slicing the whole batch — so a failed run is resumable;
  • a batch of one equals that station inside a batch.

The GIL is released for the whole batch, so it parallelises under a dask thread pool. There is deliberately no internal thread pool: two schedulers competing for the same cores oversubscribe them.

Tests

pytest                     batch invariants, seeding properties, the stub, doctests
cargo test --workspace     properties, kernel goldens, end-to-end snapshot
cargo clippy --workspace   zero warnings, enforced in CI
cargo bench                per-fault-size timings (HB_BENCH_SLOW=1 for alpine)

crates/hb_high/tests/snapshot.rs pins the whole pipeline against harness/golden/snapshot.txt using a frozen draw source. It fails by design when a change alters the draw structure (count, order or generator), because such a change needs a statistical check rather than a sample comparison. Re-record with UPDATE_SNAPSHOT=1 and say why in the commit message.

CI runs the Rust tests in both debug and release, which must agree with each other: a disagreement means the output depends on optimisation-level float behaviour.

uvx lefthook install sets up git hooks that mirror CI: autofixes on commit, and the lint, format, type and docstring gates on push.

Scope

Python owns every file format; read .stoch slip models and velocity models with source_modelling.stoch.StochFile and workflow.realisations.HFVelocityModel1D. EMOD3D itself is not modified by this project.

Metadata

Release files for hf-simulation 2026.10.1

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

Source distribution (sdist)

Source distribution for hf-simulation 2026.10.1
File Size Uploaded
hf_simulation-2026.10.1.tar.gz 893.7 kB Details

Built distributions (wheels)

Table of built distributions (wheels) for hf-simulation 2026.10.1
File
hf_simulation-2026.10.1-cp314-cp314t-win_amd64.whl CPython 3.14 CPython 3.14 free-threading Windows x86-64 Details
hf_simulation-2026.10.1-cp314-cp314t-manylinux_2_28_x86_64.whl CPython 3.14 CPython 3.14 free-threading Linux glibc 2.28+ x86-64 Details
hf_simulation-2026.10.1-cp314-cp314t-macosx_11_0_x86_64.whl CPython 3.14 CPython 3.14 free-threading macOS 11.0+ x86-64 Details
hf_simulation-2026.10.1-cp314-cp314t-macosx_11_0_arm64.whl CPython 3.14 CPython 3.14 free-threading macOS 11.0+ ARM64 Details
hf_simulation-2026.10.1-cp314-cp314-win_amd64.whl CPython 3.14 CPython 3.14 Windows x86-64 Details
hf_simulation-2026.10.1-cp314-cp314-manylinux_2_28_x86_64.whl CPython 3.14 CPython 3.14 Linux glibc 2.28+ x86-64 Details
hf_simulation-2026.10.1-cp314-cp314-macosx_11_0_x86_64.whl CPython 3.14 CPython 3.14 macOS 11.0+ x86-64 Details
hf_simulation-2026.10.1-cp314-cp314-macosx_11_0_arm64.whl CPython 3.14 CPython 3.14 macOS 11.0+ ARM64 Details
hf_simulation-2026.10.1-cp313-cp313-win_amd64.whl CPython 3.13 CPython 3.13 Windows x86-64 Details
hf_simulation-2026.10.1-cp313-cp313-manylinux_2_28_x86_64.whl CPython 3.13 CPython 3.13 Linux glibc 2.28+ x86-64 Details
hf_simulation-2026.10.1-cp313-cp313-macosx_11_0_x86_64.whl CPython 3.13 CPython 3.13 macOS 11.0+ x86-64 Details
hf_simulation-2026.10.1-cp313-cp313-macosx_11_0_arm64.whl CPython 3.13 CPython 3.13 macOS 11.0+ ARM64 Details
hf_simulation-2026.10.1-cp312-cp312-win_amd64.whl CPython 3.12 CPython 3.12 Windows x86-64 Details
hf_simulation-2026.10.1-cp312-cp312-manylinux_2_28_x86_64.whl CPython 3.12 CPython 3.12 Linux glibc 2.28+ x86-64 Details
hf_simulation-2026.10.1-cp312-cp312-macosx_11_0_x86_64.whl CPython 3.12 CPython 3.12 macOS 11.0+ x86-64 Details
hf_simulation-2026.10.1-cp312-cp312-macosx_11_0_arm64.whl CPython 3.12 CPython 3.12 macOS 11.0+ ARM64 Details

Total release size: 33.2 MB

Release files / hf_simulation-2026.10.1.tar.gz

Download URL hf_simulation-2026.10.1.tar.gz
Size 893.7 kB
Tags Source
SHA-256 checksum
How to use checksums
316492021a1cdc21610dc5e16a25691219aa265e48574dc8cd589c8cf1cc9b9f
BLAKE2b-256 checksum
How to use checksums
563e20d8f664966b6f055ff1b8ec52eab4b7addbfb0ada107e54840bfc17cc6e
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

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 Oct 12, 2026.

Transparency log

Release files / hf_simulation-2026.10.1-cp314-cp314t-win_amd64.whl

Download URL hf_simulation-2026.10.1-cp314-cp314t-win_amd64.whl
Size 694.0 kB
Tags CPython 3.14 CPython 3.14 free-threading Windows x86-64
SHA-256 checksum
How to use checksums
16e3f859853d6cd14b10ddf27c246f0f7e39473e260c2720121c3aca0245103b
BLAKE2b-256 checksum
How to use checksums
577f6ea2a3c05706cb71030003f5c08f9f3b547c9353148df0da669d33095fba
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

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 Oct 12, 2026.

Transparency log

Release files / hf_simulation-2026.10.1-cp314-cp314t-manylinux_2_28_x86_64.whl

Download URL hf_simulation-2026.10.1-cp314-cp314t-manylinux_2_28_x86_64.whl
Size 6.2 MB
Tags CPython 3.14 CPython 3.14 free-threading Linux glibc 2.28+ x86-64
SHA-256 checksum
How to use checksums
2061fc07e307106a6c6c0e06bbecc8ce6bba1cd5cf759377f3754ca5c958af02
BLAKE2b-256 checksum
How to use checksums
72b23feccad1b7af49f01a4ed8771104caf7b67556177a041beb80b382490097
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

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 Oct 12, 2026.

Transparency log

Release files / hf_simulation-2026.10.1-cp314-cp314t-macosx_11_0_x86_64.whl

Download URL hf_simulation-2026.10.1-cp314-cp314t-macosx_11_0_x86_64.whl
Size 683.4 kB
Tags CPython 3.14 CPython 3.14 free-threading macOS 11.0+ x86-64
SHA-256 checksum
How to use checksums
3576e793d858ed74af97136bcb9c508d046803e02ab557e4337cefa9f09c105c
BLAKE2b-256 checksum
How to use checksums
f24089eddf3a54d777ea59cae21c61fb0330db214bbd64ac5b0fd0db6b75a38a
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

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 Oct 12, 2026.

Transparency log

Release files / hf_simulation-2026.10.1-cp314-cp314t-macosx_11_0_arm64.whl

Download URL hf_simulation-2026.10.1-cp314-cp314t-macosx_11_0_arm64.whl
Size 478.1 kB
Tags CPython 3.14 CPython 3.14 free-threading macOS 11.0+ ARM64
SHA-256 checksum
How to use checksums
40008668d7131b5396dbb1b68bee9896c757b7bc043c7656d92169f4dc235e9c
BLAKE2b-256 checksum
How to use checksums
e296648653d9151f82c19a67108a810edf9c1c7aa40287aed6f71c2a7794a1d9
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

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 Oct 12, 2026.

Transparency log

Release files / hf_simulation-2026.10.1-cp314-cp314-win_amd64.whl

Download URL hf_simulation-2026.10.1-cp314-cp314-win_amd64.whl
Size 697.6 kB
Tags CPython 3.14 Windows x86-64
SHA-256 checksum
How to use checksums
0cb1c07636b97877ec1c5a489daa1913c81bd381c560b7b0b0d0743be90418e8
BLAKE2b-256 checksum
How to use checksums
0bf59761fad4b75f584223cccc9833ff27f006214627362db7d13d8a60e66951
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

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 Oct 12, 2026.

Transparency log

Release files / hf_simulation-2026.10.1-cp314-cp314-manylinux_2_28_x86_64.whl

Download URL hf_simulation-2026.10.1-cp314-cp314-manylinux_2_28_x86_64.whl
Size 6.2 MB
Tags CPython 3.14 Linux glibc 2.28+ x86-64
SHA-256 checksum
How to use checksums
0b94125d61a93678831d57f663fc203c95ca56c18adaa29a6a483f8da719f43a
BLAKE2b-256 checksum
How to use checksums
1128d5759ab2f4406e1bc620ccf73e79b7ce2d9c5f722b9bd0f1223a78addd9f
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

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 Oct 12, 2026.

Transparency log

Release files / hf_simulation-2026.10.1-cp314-cp314-macosx_11_0_x86_64.whl

Download URL hf_simulation-2026.10.1-cp314-cp314-macosx_11_0_x86_64.whl
Size 690.3 kB
Tags CPython 3.14 macOS 11.0+ x86-64
SHA-256 checksum
How to use checksums
c3f0d7f254efc1fec4f871de011f4adef2809beebc95be1e1ab790999e468918
BLAKE2b-256 checksum
How to use checksums
35431d6489f9b1d4c40b09a52bafe19d114dc8c5b999fdaf871512d6e430b460
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

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 Oct 12, 2026.

Transparency log

Release files / hf_simulation-2026.10.1-cp314-cp314-macosx_11_0_arm64.whl

Download URL hf_simulation-2026.10.1-cp314-cp314-macosx_11_0_arm64.whl
Size 483.0 kB
Tags CPython 3.14 macOS 11.0+ ARM64
SHA-256 checksum
How to use checksums
b7b2f06328204927e2c696bc3a10f329eae5308e01f3cb74dff58c683cd39c6c
BLAKE2b-256 checksum
How to use checksums
bf3e0edd94f5d6bd1cc9c00ef86ea67f8cedad2d4a801e3d96b8fbafee0b2c69
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

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 Oct 12, 2026.

Transparency log

Release files / hf_simulation-2026.10.1-cp313-cp313-win_amd64.whl

Download URL hf_simulation-2026.10.1-cp313-cp313-win_amd64.whl
Size 680.4 kB
Tags CPython 3.13 Windows x86-64
SHA-256 checksum
How to use checksums
535d585d92433c1165b18a46639c08f3343569681dbb86f533d750ed1d71de2e
BLAKE2b-256 checksum
How to use checksums
92e442201ce2893091a62603e738a0f7a69e8e22bbca0d6875f5b179d814343a
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

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 Oct 12, 2026.

Transparency log

Release files / hf_simulation-2026.10.1-cp313-cp313-manylinux_2_28_x86_64.whl

Download URL hf_simulation-2026.10.1-cp313-cp313-manylinux_2_28_x86_64.whl
Size 6.2 MB
Tags CPython 3.13 Linux glibc 2.28+ x86-64
SHA-256 checksum
How to use checksums
82cc66bddf314700796fa172f643db1cc750bc0bcd8ecf2a1a41a167cb4a9e5a
BLAKE2b-256 checksum
How to use checksums
faec681b97d854683acc8c6e92662ce4bc1685c21ebc9868400631b33dbd20cb
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

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 Oct 12, 2026.

Transparency log

Release files / hf_simulation-2026.10.1-cp313-cp313-macosx_11_0_x86_64.whl

Download URL hf_simulation-2026.10.1-cp313-cp313-macosx_11_0_x86_64.whl
Size 689.1 kB
Tags CPython 3.13 macOS 11.0+ x86-64
SHA-256 checksum
How to use checksums
604fa4d77573dd49d3640dc7a8baf7e4ef35bf7fb21647081f11c47db38d1f95
BLAKE2b-256 checksum
How to use checksums
ef5b00c462289e830a518a88af200b81dc1ea53346a7dc41b67fea36d474fe28
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

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 Oct 12, 2026.

Transparency log

Release files / hf_simulation-2026.10.1-cp313-cp313-macosx_11_0_arm64.whl

Download URL hf_simulation-2026.10.1-cp313-cp313-macosx_11_0_arm64.whl
Size 482.4 kB
Tags CPython 3.13 macOS 11.0+ ARM64
SHA-256 checksum
How to use checksums
bc5317bbe8aa4faac4c357400b740cab34e37c872c480e9b16a3a2108d984aaa
BLAKE2b-256 checksum
How to use checksums
30bd2c0f15aae477db2325d761c6a1ad2670930e45a0e8f271445ccabfd4e831
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

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 Oct 12, 2026.

Transparency log

Release files / hf_simulation-2026.10.1-cp312-cp312-win_amd64.whl

Download URL hf_simulation-2026.10.1-cp312-cp312-win_amd64.whl
Size 680.2 kB
Tags CPython 3.12 Windows x86-64
SHA-256 checksum
How to use checksums
48e6783c4839c1e51c0a3aa579e5beaa2ca71790fe7337e0fade9f01720e671f
BLAKE2b-256 checksum
How to use checksums
017d71b206f030786d6319e137436e990b352509ef56d6e59f8cfd4998cab429
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

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 Oct 12, 2026.

Transparency log

Release files / hf_simulation-2026.10.1-cp312-cp312-manylinux_2_28_x86_64.whl

Download URL hf_simulation-2026.10.1-cp312-cp312-manylinux_2_28_x86_64.whl
Size 6.2 MB
Tags CPython 3.12 Linux glibc 2.28+ x86-64
SHA-256 checksum
How to use checksums
f273d3ec8515073df62eda722ab4f9fd62a52c2ad2323b0e2d5d318667eda70c
BLAKE2b-256 checksum
How to use checksums
ca7d35ed5e7f4dc707e3345f3b54d35823ea664ccd671e9a06800d85eeae971d
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

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 Oct 12, 2026.

Transparency log

Release files / hf_simulation-2026.10.1-cp312-cp312-macosx_11_0_x86_64.whl

Download URL hf_simulation-2026.10.1-cp312-cp312-macosx_11_0_x86_64.whl
Size 688.9 kB
Tags CPython 3.12 macOS 11.0+ x86-64
SHA-256 checksum
How to use checksums
4602532f4e502bd831c4c3aa7208ec0667b7d887acdc09af2094eb4af3b116b6
BLAKE2b-256 checksum
How to use checksums
892dd3d890d71588fcb95f50c7d10e04fc62cdb3a612bcc3d5d91c4de04bd301
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

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 Oct 12, 2026.

Transparency log

Release files / hf_simulation-2026.10.1-cp312-cp312-macosx_11_0_arm64.whl

Download URL hf_simulation-2026.10.1-cp312-cp312-macosx_11_0_arm64.whl
Size 482.2 kB
Tags CPython 3.12 macOS 11.0+ ARM64
SHA-256 checksum
How to use checksums
8c102bc2c7c7936740b52831879757cd1bdebf4a84db81ab1cbbfc86b13a1a26
BLAKE2b-256 checksum
How to use checksums
c284514993a3cab265be3ab1b2efc9dfeec18e575b552c6d5b56f571e8f8be15
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

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 Oct 12, 2026.

Transparency log

Release history Release notifications | RSS feed

This release

2026.10.1 This release

17 release files

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