lazybgen

CI Wheels PyPI License: MIT

High-performance BGEN reader with Google Cloud Storage and Amazon S3 partial-read support. lazybgen reads only the variants or regions you ask for, fetching them directly from local files, GCS, or S3 via random-access byte-range reads, so there is no need to download the whole file.

It is a Cython/C++ implementation with vendored, optimized compression backends (libdeflate and zstd) compiled from source for consistent cross-platform behavior and speed (SIMD genotype parsing, parallel block decompression).

Install

pip install lazybgen

For Amazon S3 support, install the s3 extra:

pip install lazybgen[s3]

From source

git clone --recursive https://github.com/mkanai/lazybgen.git
cd lazybgen
pip install .

Building from source requires a C/C++ compiler (CMake is needed on Windows only). The vendored libdeflate and zstd are git submodules, so clone with --recursive (or run git submodule update --init --recursive).

Usage

from lazybgen import load_bgen

# Local file
genotypes, variant_info, sample_ids = load_bgen(
    "chr1.bgen",
    region="chr1:1000000-2000000",   # partial read: only this region is fetched
)

# GCS (default credentials)
load_bgen("gs://bucket/file.bgen", index_path="gs://bucket/file.bgen.bgi")

# GCS requester-pays bucket. True bills the default project from your environment
load_bgen("gs://bucket/file.bgen", storage_options={"requester_pays": True})
# ...or pass a project id string to bill a specific project
load_bgen("gs://bucket/file.bgen", storage_options={"requester_pays": "my-billing-project"})

# Public S3 bucket (anonymous, no credentials)
load_bgen("s3://bucket/file.bgen", storage_options={"anon": True})

gs:// reads go through obstore by default, which is 1.1x to 3.5x faster than gcsfs and needs no code change. See Remote transports.

load_bgen returns (genotypes, variant_info, sample_ids), where genotypes is an (n_samples, n_variants) np.ndarray, variant_info is a pd.DataFrame with columns chrom, pos, rsid, ref, alt, and sample_ids is a list[str].

A .bgi index is required; create one with bgenix -g file.bgen.

Parameters

  • file_path: path, gs://, or s3:// URL to the BGEN file
  • index_path: .bgi index (defaults to file_path + ".bgi")
  • sample_path: optional .sample file
  • region: "chr:start-end" to read a genomic interval
  • variant_filter: variant subset as a dict with keys chromosome, positions, allele1 and allele2, the three lists aligned element by element and the alleles matching the file exactly (see Reading many variants); build it with load_variant_filter("variants.z"), which reads variant IDs/positions from a .z file (from lazybgen import load_variant_filter)
  • sample_ids: subset of samples to load
  • dtype: dosage dtype (default float64). np.float32 decodes ~18% faster and uses half the memory; use it when single precision is sufficient (dosages are computed in single precision regardless, so float64 output is the exact widening of the float32 result)
  • show_progress: show a progress bar while loading (default False)
  • nan_action: how to handle missing dosages: "error" (default, raise), "mean" (impute with the per-variant mean), "omit" (drop affected samples), or "warn" (keep NaNs and log a warning)
  • num_threads: worker threads for decoding. 0 (default) auto-detects the CPU core count and decodes blocks in parallel; 1 forces single-threaded decoding; N > 1 uses N threads (see Parallel decode)
  • storage_options: cloud backend kwargs, in fsspec spelling whichever transport serves them (e.g. {"anon": True} for public S3, {"requester_pays": True} to bill the env default project, or {"requester_pays": "billing-project-id"} for GCS requester-pays buckets)
  • remote_backend: transport for gs:// and s3:// reads: "auto" (default), "obstore", or "fsspec" (see Remote transports)

Supported BGEN features

lazybgen computes alt-allele dosages and targets the most common BGEN profile:

Feature Support
Layout v1.2 / v1.3 Yes
Layout v1.1 Best-effort
Compression: zlib, zstd Yes
Compression: none (uncompressed) No
Biallelic, diploid (phased or unphased) Yes
Multi-allelic (>2 alleles) No
Non-diploid (ploidy != 2) No

Unsupported inputs raise a clear error rather than returning wrong dosages. Compress uncompressed files with bgenix or qctool2 first.

Layout v1.1 is an older format with a different probability encoding; lazybgen decodes it through a separate, less-exercised path, so it is best-effort. Prefer v1.2 / v1.3 (re-encode with qctool2 if needed) for production use.

Reading many variants

Ask for the variants together rather than one call each. Every call returns a variant-info DataFrame, and pandas charges about 100 us to build one whatever its row count, so a loop of point lookups spends more time in pandas than in reading. For 500 variants the batched forms below are ~17x faster than the loop:

from lazybgen import load_bgen
from lazybgen.reader import BgenReader

# Your variants: positions and their alleles, aligned element by element.
positions = [10_001, 25_500, 91_200]  # ...hundreds more
ref_alleles = ["A", "C", "G"]
alt_alleles = ["G", "T", "A"]

# Slow: one call, one DataFrame, per variant.
for pos in positions:
    load_bgen("chr1.bgen", region=f"chr1:{pos}-{pos}")

# Fast: one call for the whole selection.
genotypes, variant_info, sample_ids = load_bgen(
    "chr1.bgen",
    variant_filter={
        "chromosome": "chr1",
        "positions": positions,
        "allele1": ref_alleles,
        "allele2": alt_alleles,
    },
)

# Also fast, and memory-bounded: stream a contiguous range.
with BgenReader("chr1.bgen") as reader:
    for info, dosage in reader.iter_variants(region_chrom="chr1", region_start=10_000, region_end=100_000):
        ...

All four keys are required, and the alleles must match the file exactly: a variant whose alleles differ is not matched, and neither is one whose ref/alt are the other way round, so a filter that matches nothing raises rather than returning an empty result. load_variant_filter("variants.z") builds the dict from a .z file if your variants come from one.

Memory

A local file is memory-mapped and read in place, so the compressed bytes are never copied on the way and a read costs about the matrix it returns. lazybgen peaks around 1.2x the bgen package for the same read (9.7 GB against 7.8 GB for a 2000-variant region at 500k samples).

Ask for float32 when single precision is enough. Dosages are computed in single precision either way, so a float64 result is the exact widening of the same numbers, and asking for it costs twice the memory and ~18% more decode time:

import numpy as np

genotypes, _, _ = load_bgen("chr1.bgen", region="chr1:1-1000000", dtype=np.float32)

To avoid materializing the matrix at all, stream it: iter_variants is O(n_samples x block_size) whatever the file holds (see Streaming large files).

Streaming large files

load_bgen materializes the whole (n_samples, n_variants) matrix. For files too large to hold at once, BgenReader.iter_variants() streams variants in memory-bounded blocks (peak memory O(n_samples x block_size)). It yields (info, dosage) per variant, where info is a dict with keys chrom, pos, rsid, ref, alt (access as info["chrom"]) and dosage is a 1-D array of per-sample dosages (NaN for missing):

from lazybgen.reader import BgenReader

with BgenReader("chr1.bgen") as reader:
    for info, dosage in reader.iter_variants():
        ...  # info["pos"]; dosage.shape == (reader.nsamples,)

It accepts the same region_chrom/region_start/region_end, variant_filter, sample_indices, and dtype selection as load_variants. By default block_size auto-scales to keep each block near a fixed memory budget (it shrinks as the sample count grows); pass an explicit block_size to override.

Parallel decode

Decoding runs in parallel across CPU cores by default: each block is inflated and decoded across worker threads, byte-identical to single-threaded decoding. This applies to load_bgen, BgenReader.load_variants, and iter_variants, for both all-samples and sample-filtered (cohort) reads, and scales with the sample and core count: on a 16-core machine at 100k samples, a full decode is 6.8x faster than num_threads=1 and a 2000-variant region 6.6x.

Control the worker count with num_threads: 0 (default) auto-detects the core count, 1 forces single-threaded decoding, and N > 1 uses N threads.

# Parallel by default (auto-detected cores)
genotypes, variant_info, sample_ids = load_bgen("chr1.bgen")

# Force single-threaded decoding
load_bgen("chr1.bgen", num_threads=1)

# BgenReader takes the same num_threads control
with BgenReader("chr1.bgen", num_threads=8) as reader:
    dosages, info = reader.load_variants(region_chrom="chr1", region_start=1, region_end=1_000_000)

Remote transports

gs:// reads go through obstore, which does HTTP and TLS in Rust with the GIL released, so many range requests are genuinely in flight at once. s3:// goes through fsspec (s3fs), because obstore's S3 store assumes the us-east-1 region and does not read ~/.aws/credentials, AWS_PROFILE or SSO. obstore and gcsfs are installed by default; s3:// additionally needs pip install lazybgen[s3]. storage_options are spelled the same way for either transport, and requester-pays works on both.

remote_backend overrides the choice per reader ("auto", "obstore", "fsspec"), and LAZYBGEN_REMOTE_BACKEND does it for a process. "auto" falls back to fsspec whenever obstore is unavailable or cannot express one of your storage_options, so an option that decides which bytes come back is never silently dropped.

Against gcsfs, obstore is worth 1.1x to 3.5x end to end depending on the read: most for small ones, which are mostly connection and request latency, and least for a full decode, which is bandwidth plus decode.

Multiprocessing: obstore's runtime does not survive fork(), the default start method on Linux. Children of a process that has already read fall back to fsspec automatically rather than hanging; use the spawn or forkserver start method to keep the faster transport in workers.

Remote .bgi index caching

For a gs:///s3:// BGEN, the genotype data is read in place via byte ranges, but the .bgi index is downloaded once to a local cache. The cache lives in a dedicated directory (a lazybgen-bgi-cache subdirectory of the system temp dir) and each entry is keyed by a hash of the full URL, so same-named indexes in different buckets never collide. Override the location with the LAZYBGEN_BGI_CACHE_DIR environment variable.

Build info

from lazybgen import get_build_info returns the compression backend the package was built against (vendored libdeflate / zstd, or system libraries).

Performance

Against the bgen package on the same local files, 10k variants and samples scaling to biobank size (16 vCPU n2 VM, median of 3 warm runs, lazybgen's wall time in parentheses):

Workload 5k samples (94 MB) 50k (931 MB) 500k (9.1 GB)
Full decode 10.7x (75 ms) 15.3x (483 ms) 15.5x (4.80 s)
Region (2000 variants) 8.8x (18 ms) 15.3x (99 ms) 16.5x (961 ms)
Scattered (1000 variants) 6.0x (15 ms) 13.5x (57 ms) 15.5x (501 ms)

Remote: lazy partial reads at biobank scale

This is the point of the package. A local-only reader must download the whole file before reading a byte, so its cost tracks file size; lazybgen fetches only the variants you ask for, so its cost tracks the slice and does not grow with the file. Each cell is the end-to-end speedup over downloading the file first (whole-file gcloud storage cp + a local read), with that baseline in parentheses:

Read (500k samples) lazybgen gs:// 10k var (9.1 GB) 50k var (45 GB) 100k var (91 GB)
One variant 341 ms ~15x (5.2 s) ~76x (26 s) ~151x (52 s)
Region (2000 contiguous) 5.93 s 0.9x (5.2 s) ~4x (26 s) ~9x (52 s)
Scattered (1000 random) 3.12 s ~2x (5.2 s) ~8x (26 s) ~17x (52 s)

Only the 9.1 GB download is measured (5.2 s, median of 5); the 26 s and 52 s figures scale it by byte size, since download time depends on bytes and not on what is in them.

The lazybgen column is one number per row because it does not change with the file: the same read costs the same at 10k variants and at 100k. The 0.9x is the honest edge of that: at 10k variants a 2000-variant region is a fifth of the file, and fetching a fifth of an object costs about what fetching all of it does on a fast same-region link. Partial reads pay when the slice is genuinely a slice, which is the regime biobank-scale files are in. The baseline is also best-case for the download - same-region, free egress, and not very repeatable run to run - so a laptop, cross-region or metered link widens every gap.

These are single batched reads. A loop of one-variant calls is ~17x slower for the same variants, and a read costs about the matrix it returns; see Reading many variants and Memory.

benchmarks/README.md has the full size ladder, remote and transport comparisons, peak-memory tables, and the methodology.

License

MIT

Citation

Kanai, M. et al. Population-scale multiome immune cell atlas reveals complex disease drivers. medRxiv (2025)

Contact

Masahiro Kanai (mkanai@broadinstitute.org)

Project links

Data verified by PyPI on Sep 5, 2026
Data provided by the project maintainers, verified at the time the release was uploaded to PyPI.

Key dates

PyPI data
Data sourced directly from PyPI's database.
  • Released:
Latest release

1 maintainer

PyPI data
Data sourced directly from PyPI's database.
Avatar for mkanai from gravatar.com mkanai

Credits

Author: Masahiro Kanai

GitHub Statistics

Data verified by PyPI on Sep 5, 2026
The GitHub source repository was provided by the project maintainers and verified by PyPI at the time of upload. Stars, forks, and open issues/PRs are derived from that repository and have not been independently verified.

License expression

MIT
View SPDX License List

Requires

Python >=3.10

Provides Extra

s3 obstore dev

Tags

bgen genetics genomics gwas bioinformatics dosage gcs s3

Classifiers

Report project as malware

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

lazybgen-0.2.0.tar.gz (4.9 MB view details)

Uploaded Source

Built Distributions

If you're not sure about the file name format, learn more about wheel file names.

lazybgen-0.2.0-cp314-cp314-musllinux_1_2_x86_64.whl (4.8 MB view details)

Uploaded CPython 3.14musllinux: musl 1.2+ x86-64

lazybgen-0.2.0-cp314-cp314-manylinux_2_28_x86_64.whl (3.6 MB view details)

Uploaded CPython 3.14manylinux: glibc 2.28+ x86-64

lazybgen-0.2.0-cp314-cp314-macosx_11_0_arm64.whl (335.3 kB view details)

Uploaded CPython 3.14macOS 11.0+ ARM64

lazybgen-0.2.0-cp313-cp313-musllinux_1_2_x86_64.whl (4.8 MB view details)

Uploaded CPython 3.13musllinux: musl 1.2+ x86-64

lazybgen-0.2.0-cp313-cp313-manylinux_2_28_x86_64.whl (3.6 MB view details)

Uploaded CPython 3.13manylinux: glibc 2.28+ x86-64

lazybgen-0.2.0-cp313-cp313-macosx_11_0_arm64.whl (334.4 kB view details)

Uploaded CPython 3.13macOS 11.0+ ARM64

lazybgen-0.2.0-cp312-cp312-musllinux_1_2_x86_64.whl (4.8 MB view details)

Uploaded CPython 3.12musllinux: musl 1.2+ x86-64

lazybgen-0.2.0-cp312-cp312-manylinux_2_28_x86_64.whl (3.6 MB view details)

Uploaded CPython 3.12manylinux: glibc 2.28+ x86-64

lazybgen-0.2.0-cp312-cp312-macosx_11_0_arm64.whl (334.7 kB view details)

Uploaded CPython 3.12macOS 11.0+ ARM64

lazybgen-0.2.0-cp311-cp311-musllinux_1_2_x86_64.whl (4.9 MB view details)

Uploaded CPython 3.11musllinux: musl 1.2+ x86-64

lazybgen-0.2.0-cp311-cp311-manylinux_2_28_x86_64.whl (3.6 MB view details)

Uploaded CPython 3.11manylinux: glibc 2.28+ x86-64

lazybgen-0.2.0-cp311-cp311-macosx_11_0_arm64.whl (332.8 kB view details)

Uploaded CPython 3.11macOS 11.0+ ARM64

lazybgen-0.2.0-cp310-cp310-musllinux_1_2_x86_64.whl (4.8 MB view details)

Uploaded CPython 3.10musllinux: musl 1.2+ x86-64

lazybgen-0.2.0-cp310-cp310-manylinux_2_28_x86_64.whl (3.6 MB view details)

Uploaded CPython 3.10manylinux: glibc 2.28+ x86-64

lazybgen-0.2.0-cp310-cp310-macosx_11_0_arm64.whl (334.0 kB view details)

Uploaded CPython 3.10macOS 11.0+ ARM64

File details

Details for the file lazybgen-0.2.0.tar.gz.

File metadata

  • Download URL: lazybgen-0.2.0.tar.gz
  • Upload date:
  • Size: 4.9 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for lazybgen-0.2.0.tar.gz
Algorithm Hash digest
SHA256 71f65d704b7326da55c45c63d9a956acb55b4b07fc07f5e6e4ac126ab916ab3e
MD5 180e2c257f1519eb29f546846d9f8f7c
BLAKE2b-256 42d654bfb6a83b666e806287945a687c7db085f8a0d68bbe8a243ae16bdf706f

See more details on using hashes here.

Provenance

The following attestation bundles were made for lazybgen-0.2.0.tar.gz:

Publisher: wheels.yml on mkanai/lazybgen

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file lazybgen-0.2.0-cp314-cp314-musllinux_1_2_x86_64.whl.

File metadata

File hashes

Hashes for lazybgen-0.2.0-cp314-cp314-musllinux_1_2_x86_64.whl
Algorithm Hash digest
SHA256 517d5d6f8c74235b69ca75dbdd7e673a1b0228883178885d2ccaff8b3b81d389
MD5 6a1c14dd0b3356b8d0a098b381cd5dbd
BLAKE2b-256 13401da7e15c963577bcc1940f0156778682c2ee184185eac9c2d988f7d4f9d9

See more details on using hashes here.

Provenance

The following attestation bundles were made for lazybgen-0.2.0-cp314-cp314-musllinux_1_2_x86_64.whl:

Publisher: wheels.yml on mkanai/lazybgen

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file lazybgen-0.2.0-cp314-cp314-manylinux_2_28_x86_64.whl.

File metadata

File hashes

Hashes for lazybgen-0.2.0-cp314-cp314-manylinux_2_28_x86_64.whl
Algorithm Hash digest
SHA256 24721426d66484f41debb131e5a0c54ee5da8284a1522ab48f451d95ddf541bc
MD5 1e8241e0d4ccff1826b8e9372af4f5c2
BLAKE2b-256 938a399443819ea3a9d49e84425ddcbede5098ab155cb73aa2401f697b06e3c4

See more details on using hashes here.

Provenance

The following attestation bundles were made for lazybgen-0.2.0-cp314-cp314-manylinux_2_28_x86_64.whl:

Publisher: wheels.yml on mkanai/lazybgen

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file lazybgen-0.2.0-cp314-cp314-macosx_11_0_arm64.whl.

File metadata

File hashes

Hashes for lazybgen-0.2.0-cp314-cp314-macosx_11_0_arm64.whl
Algorithm Hash digest
SHA256 6ff5d9e0a3d8f9f7ad299c4a23a27f092fef9c52ca89d519bca9ce17ac0ac4c7
MD5 997c81dfbcea148f77d959071fb99cd5
BLAKE2b-256 13fbb752070f33a3b620156f6f8697acb22d3d1966b5b425fea9201243de387c

See more details on using hashes here.

Provenance

The following attestation bundles were made for lazybgen-0.2.0-cp314-cp314-macosx_11_0_arm64.whl:

Publisher: wheels.yml on mkanai/lazybgen

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file lazybgen-0.2.0-cp313-cp313-musllinux_1_2_x86_64.whl.

File metadata

File hashes

Hashes for lazybgen-0.2.0-cp313-cp313-musllinux_1_2_x86_64.whl
Algorithm Hash digest
SHA256 2c0e7a857b130d17f2d2bc9349eecb88608d78f1c2e6ba44c72a972e11e7d52a
MD5 fb946bb7aed9cf2a2d1f76556bc027ee
BLAKE2b-256 c37378bc886896757a47b67869f15ff87a1182352f3a157e5a50a976377132d2

See more details on using hashes here.

Provenance

The following attestation bundles were made for lazybgen-0.2.0-cp313-cp313-musllinux_1_2_x86_64.whl:

Publisher: wheels.yml on mkanai/lazybgen

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file lazybgen-0.2.0-cp313-cp313-manylinux_2_28_x86_64.whl.

File metadata

File hashes

Hashes for lazybgen-0.2.0-cp313-cp313-manylinux_2_28_x86_64.whl
Algorithm Hash digest
SHA256 90509626586175536b2e975d1198fe8697cc875f4a1912f1366317210c083eb4
MD5 98ac256b0a8523cc72bc137dfbffd40e
BLAKE2b-256 b57fde04c0e88f5461352701183824e7981544071b0d14a74ddf007568626d0e

See more details on using hashes here.

Provenance

The following attestation bundles were made for lazybgen-0.2.0-cp313-cp313-manylinux_2_28_x86_64.whl:

Publisher: wheels.yml on mkanai/lazybgen

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file lazybgen-0.2.0-cp313-cp313-macosx_11_0_arm64.whl.

File metadata

File hashes

Hashes for lazybgen-0.2.0-cp313-cp313-macosx_11_0_arm64.whl
Algorithm Hash digest
SHA256 c80b920e451281058a287fd9f3cc3244438ab06e246ba9be223c42dfa3f5d1bc
MD5 fd4b7a77366fce44f655b437706328b1
BLAKE2b-256 945757dec32372a681285624983608739571be6afc640e5d7954beb70b04afae

See more details on using hashes here.

Provenance

The following attestation bundles were made for lazybgen-0.2.0-cp313-cp313-macosx_11_0_arm64.whl:

Publisher: wheels.yml on mkanai/lazybgen

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file lazybgen-0.2.0-cp312-cp312-musllinux_1_2_x86_64.whl.

File metadata

File hashes

Hashes for lazybgen-0.2.0-cp312-cp312-musllinux_1_2_x86_64.whl
Algorithm Hash digest
SHA256 4d1f2530182755cccd60b8eb6b326d1b7429a9ef57d7636b6691beb9225e9130
MD5 fb1792d8c7653ef4b8cf94124dd9690e
BLAKE2b-256 c64464383f364152192be0779496e43a820563380c55b07bdfc34358265bb1b8

See more details on using hashes here.

Provenance

The following attestation bundles were made for lazybgen-0.2.0-cp312-cp312-musllinux_1_2_x86_64.whl:

Publisher: wheels.yml on mkanai/lazybgen

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file lazybgen-0.2.0-cp312-cp312-manylinux_2_28_x86_64.whl.

File metadata

File hashes

Hashes for lazybgen-0.2.0-cp312-cp312-manylinux_2_28_x86_64.whl
Algorithm Hash digest
SHA256 4d1683e419277ebb7cfcc983327e604a86a6f7f2c2bb216e1a011c8b782e625c
MD5 6c1328c716156ba30cedc746a2f59dda
BLAKE2b-256 27843ac511fc98c0f9f390bcede699372f14c1c7bc2f5df0da9c7059e1f5b50f

See more details on using hashes here.

Provenance

The following attestation bundles were made for lazybgen-0.2.0-cp312-cp312-manylinux_2_28_x86_64.whl:

Publisher: wheels.yml on mkanai/lazybgen

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file lazybgen-0.2.0-cp312-cp312-macosx_11_0_arm64.whl.

File metadata

File hashes

Hashes for lazybgen-0.2.0-cp312-cp312-macosx_11_0_arm64.whl
Algorithm Hash digest
SHA256 95bd77419b14e8596684387b8cf3711e8ee88b0d0ee4ac688155d107904bc51c
MD5 8f2aa88e028831cfe1cc5537c842bc46
BLAKE2b-256 23510245bcb243689ccea263fb6362d84456f0cd9540b4b00c86b5d17d0a4f4d

See more details on using hashes here.

Provenance

The following attestation bundles were made for lazybgen-0.2.0-cp312-cp312-macosx_11_0_arm64.whl:

Publisher: wheels.yml on mkanai/lazybgen

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file lazybgen-0.2.0-cp311-cp311-musllinux_1_2_x86_64.whl.

File metadata

File hashes

Hashes for lazybgen-0.2.0-cp311-cp311-musllinux_1_2_x86_64.whl
Algorithm Hash digest
SHA256 ec0ad8471bff9a4d65bc847526680bc9740bcb6cecfb0b4090c5de5112ce294e
MD5 ddabc70cb36c9c0633f686abfcba04ee
BLAKE2b-256 0a47b12d44e5109049f0bd480a2767386b373144350456ffeedbbbe95210d94a

See more details on using hashes here.

Provenance

The following attestation bundles were made for lazybgen-0.2.0-cp311-cp311-musllinux_1_2_x86_64.whl:

Publisher: wheels.yml on mkanai/lazybgen

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file lazybgen-0.2.0-cp311-cp311-manylinux_2_28_x86_64.whl.

File metadata

File hashes

Hashes for lazybgen-0.2.0-cp311-cp311-manylinux_2_28_x86_64.whl
Algorithm Hash digest
SHA256 d766b8ee0780842ea81ee4357b3cbf5f1d55afbdc5ce0bccc7370128d3505b4f
MD5 0b79133ac29388466423b6e880bb7e92
BLAKE2b-256 c8520d65e6928422d96240c3b9ec0895578e9e0f53214e7ba5ae3548bff76d5b

See more details on using hashes here.

Provenance

The following attestation bundles were made for lazybgen-0.2.0-cp311-cp311-manylinux_2_28_x86_64.whl:

Publisher: wheels.yml on mkanai/lazybgen

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file lazybgen-0.2.0-cp311-cp311-macosx_11_0_arm64.whl.

File metadata

File hashes

Hashes for lazybgen-0.2.0-cp311-cp311-macosx_11_0_arm64.whl
Algorithm Hash digest
SHA256 4975e88e057704cc97086b482680c72f5074fccaddc38514fcee5f0c0e5f148d
MD5 1656f4fef6f9db369bccd8d4c23add67
BLAKE2b-256 13abfc579e3d05bb3802f5789127f778a23016fda533bdc5c3d7169bab5251cf

See more details on using hashes here.

Provenance

The following attestation bundles were made for lazybgen-0.2.0-cp311-cp311-macosx_11_0_arm64.whl:

Publisher: wheels.yml on mkanai/lazybgen

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file lazybgen-0.2.0-cp310-cp310-musllinux_1_2_x86_64.whl.

File metadata

File hashes

Hashes for lazybgen-0.2.0-cp310-cp310-musllinux_1_2_x86_64.whl
Algorithm Hash digest
SHA256 101a0fdd64e1057689543777f18e2f3b92a49e5b2d3a0192a864f77793a87f48
MD5 1f3424cb0839efb4406d65a1b6a8c380
BLAKE2b-256 b7c4aff514c06cd845897588e5f236f72be3a6c831bcad93e1cc73c3ae120d0f

See more details on using hashes here.

Provenance

The following attestation bundles were made for lazybgen-0.2.0-cp310-cp310-musllinux_1_2_x86_64.whl:

Publisher: wheels.yml on mkanai/lazybgen

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file lazybgen-0.2.0-cp310-cp310-manylinux_2_28_x86_64.whl.

File metadata

File hashes

Hashes for lazybgen-0.2.0-cp310-cp310-manylinux_2_28_x86_64.whl
Algorithm Hash digest
SHA256 32512f9a47fa53c8ea0f0b67a27397d2cd17a3c5032f6150f0e9d4896c9cc29e
MD5 b2c250054136784c87b68c1b0bff66d5
BLAKE2b-256 1db1143e92573d3395e8eefd7221df0aa17cab5ccfce3d79cd83ee9165252c24

See more details on using hashes here.

Provenance

The following attestation bundles were made for lazybgen-0.2.0-cp310-cp310-manylinux_2_28_x86_64.whl:

Publisher: wheels.yml on mkanai/lazybgen

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file lazybgen-0.2.0-cp310-cp310-macosx_11_0_arm64.whl.

File metadata

File hashes

Hashes for lazybgen-0.2.0-cp310-cp310-macosx_11_0_arm64.whl
Algorithm Hash digest
SHA256 e73ee8267f8499de20a1040ae8c19ed3f58d4898a0ca10634d0744f66dab7c66
MD5 de7cdfe9423df824673e878d30cb7823
BLAKE2b-256 c491d015ea5ebb5d6edc169d27ea9cd2ad1c5c5eef9bd46d5562f67e5a10ea5a

See more details on using hashes here.

Provenance

The following attestation bundles were made for lazybgen-0.2.0-cp310-cp310-macosx_11_0_arm64.whl:

Publisher: wheels.yml on mkanai/lazybgen

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Release history Release notifications | RSS feed

This release

0.2.0 This release

16 files

0.1.0

19 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