Skip to main content

aion-engine

Python bindings for Aion — a graph-IR tensor runtime with a Zig core, CPU and GPU. The distribution is aion-engine; the import name is aion.

Install

pip install aion-engine          # CPU
pip install aion-engine[gpu]     # + the GPU (wgpu) runtime

Prebuilt wheels for CPython 3.12/3.13 on Windows x64, Linux x86_64/aarch64 and macOS arm64/x86_64 — no Zig or C toolchain needed. The Aion runtime is static-linked into the extension module, so there is no separate aion.dll/libaion.so to ship or locate.

The GPU backend is compiled into those same wheels but links no wgpu — it dlopens wgpu-native at runtime. [gpu] adds the companion aion-wgpu wheel, which ships that library for every platform the runtime wheels target. Without it, GPU calls raise aion.AionError and CPU is unaffected.

NumPy is optional ([numpy] extra); without it use tensor.tolist() / tensor.item().

Run a model

A .aion file carries graph, weights and metadata, so this is the whole setup:

import aion, numpy as np

model = aion.load_model("model.aion")              # device="gpu" to run on GPU
out = model.run_numpy({"x": np.zeros((1, 4), "f4")})["y"]

In hot loops, bind tensors once (model.bind_input(name, t) + model.run()) instead of creating temporaries per step. KV caches and streaming state are managed by the runtime from the model's input roles — you don't thread them through Python (load_model(..., cache_capacity=, growable=) tunes it).

Build a model

Two handles, and they don't mix. aion.Tensor is data: what you write inputs into, read outputs from, and hand a layer as a weight. aion.TensorRef is a value in a graph: what ops and aion.nn layers take and return.

import aion, numpy as np
from aion import nn

ctx = aion.Context()
w = aion.tensor(np.random.randn(3, 4).astype("f4"))   # data, [out, in]

with aion.Builder(ctx) as b:
    x = b.input((1, 4)).rename("x")                   # TensorRef
    y = nn.Linear(w)(x).relu().rename("y")
    model = b.compile([y])

out = model.run_numpy({"x": np.ones((1, 4), "f4")})["y"]

For a reusable module, aion.compile(MyModule(), aion.spec((None, 4))) traces forward once (None = dynamic axis) and aion.export(..., "mlp.aion") writes the package. compile works on a copy of the graph, so authoring survives it: you can compile twice, and print(y) / y.numpy() evaluates just that value's cone while you keep building.

dtype takes Aion constants (aion.float32, float16, int8, int32, q8_0, q4_0) or NumPy dtypes — not strings.

GPU

ctx = aion.Context.gpu()                                  # first discrete adapter
model = aion.LoadedModel.load(ctx, "model.aion", device="gpu")

Keep feeding CPU input tensors: the model migrates them on run() and flushes outputs back for reading. A standalone tensor.to("gpu") is a move — the host copy is freed, so to("cpu") before reading it. device="gpu:1" (or adapter_index=) picks a physical adapter, power="low"/"high" picks integrated/discrete, and Context(gpus=[...]) registers several.

Examples

Runnable scripts in bindings/python/examples/, all taking --device {cpu,gpu,auto}:

silero_vad_simple.py chunked VAD, reports throughput
harrier_embed.py text embeddings, ranks documents against a query
nemotron_asr_streaming.py, nemotron_asr_mic.py streaming ASR from a wav or the mic
gemma4_e2b_generate_one.py LLM token generation

Models are produced by the converters in scripts/.

Developing from a checkout

uv sync        # environment + native build (dev group: numpy, torch, pytest, aion-wgpu)
uv run pytest

uv run rebuilds and relinks the extension when the Zig core changes — the uv cache key covers src/**/*.zig, the public headers and the build files. uv sync --reinstall-package aion-engine forces a rebuild.

Build knobs: AION_PY_OPTIMIZE (default ReleaseFast), AION_PY_CPU (native), AION_PY_TARGET, AION_PY_GPU (overrides [tool.aion] gpu in pyproject.toml). Building from source needs Zig on PATH plus a C toolchain.

Notes: prefer with for deterministic cleanup (Context.close() closes live tensors/models first); AION_DEFAULT_THREAD_COUNT sets the default context's thread count, and aion.reset_default_context() re-reads it.

Metadata

Release files for aion-engine 0.0.5

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

Built distributions (wheels)

Table of built distributions (wheels) for aion-engine 0.0.5
File Actions
aion_engine-0.0.5-cp313-cp313-win_amd64.whl CPython 3.13 CPython 3.13 Windows x86-64 Details for aion_engine-0.0.5-cp313-cp313-win_amd64.whl
aion_engine-0.0.5-cp313-cp313-manylinux_2_28_x86_64.whl CPython 3.13 CPython 3.13 Linux glibc 2.28+ x86-64 Details for aion_engine-0.0.5-cp313-cp313-manylinux_2_28_x86_64.whl
aion_engine-0.0.5-cp313-cp313-manylinux_2_28_aarch64.whl CPython 3.13 CPython 3.13 Linux glibc 2.28+ ARM64 Details for aion_engine-0.0.5-cp313-cp313-manylinux_2_28_aarch64.whl
aion_engine-0.0.5-cp313-cp313-macosx_11_0_arm64.whl CPython 3.13 CPython 3.13 macOS 11.0+ ARM64 Details for aion_engine-0.0.5-cp313-cp313-macosx_11_0_arm64.whl
aion_engine-0.0.5-cp312-cp312-win_amd64.whl CPython 3.12 CPython 3.12 Windows x86-64 Details for aion_engine-0.0.5-cp312-cp312-win_amd64.whl
aion_engine-0.0.5-cp312-cp312-manylinux_2_28_x86_64.whl CPython 3.12 CPython 3.12 Linux glibc 2.28+ x86-64 Details for aion_engine-0.0.5-cp312-cp312-manylinux_2_28_x86_64.whl
aion_engine-0.0.5-cp312-cp312-manylinux_2_28_aarch64.whl CPython 3.12 CPython 3.12 Linux glibc 2.28+ ARM64 Details for aion_engine-0.0.5-cp312-cp312-manylinux_2_28_aarch64.whl
aion_engine-0.0.5-cp312-cp312-macosx_11_0_arm64.whl CPython 3.12 CPython 3.12 macOS 11.0+ ARM64 Details for aion_engine-0.0.5-cp312-cp312-macosx_11_0_arm64.whl

Total release size: 27.0 MB

Release files / aion_engine-0.0.5-cp313-cp313-win_amd64.whl

Download URL aion_engine-0.0.5-cp313-cp313-win_amd64.whl
Size 1.5 MB
Tags CPython 3.13 Windows x86-64
SHA-256 checksum
How to use checksums
0b71a002f6cba51410ac1cf30b8e38802d597c84884fdbbc771ba34dd62343d0
BLAKE2b-256 checksum
How to use checksums
9f0cb736282035406b1222babab16c139c7018ac4dcbe7be45d0e02a3df3282c
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 Oct 20, 2026.

Transparency log

Release files / aion_engine-0.0.5-cp313-cp313-manylinux_2_28_x86_64.whl

Download URL aion_engine-0.0.5-cp313-cp313-manylinux_2_28_x86_64.whl
Size 6.0 MB
Tags CPython 3.13 Linux glibc 2.28+ x86-64
SHA-256 checksum
How to use checksums
a68431722baa50bff28792bc9183c6e523062e54b6b2c4baa211a2b3babea4d8
BLAKE2b-256 checksum
How to use checksums
8c89f58f10aaedd1f2ed11ea96188bcf0725cf4ec57ccb4f2e069bc3c7d56b4d
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 Oct 20, 2026.

Transparency log

Release files / aion_engine-0.0.5-cp313-cp313-manylinux_2_28_aarch64.whl

Download URL aion_engine-0.0.5-cp313-cp313-manylinux_2_28_aarch64.whl
Size 4.3 MB
Tags CPython 3.13 Linux glibc 2.28+ ARM64
SHA-256 checksum
How to use checksums
e6f53e6418e48a157b52ad770c0c5adfc7a4641b877ada720ba0334e20fcef67
BLAKE2b-256 checksum
How to use checksums
02729f1b0a566cfeee493ba05649507cc2d9c5b16d7326c23290aa3db1fc0165
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 Oct 20, 2026.

Transparency log

Release files / aion_engine-0.0.5-cp313-cp313-macosx_11_0_arm64.whl

Download URL aion_engine-0.0.5-cp313-cp313-macosx_11_0_arm64.whl
Size 1.7 MB
Tags CPython 3.13 macOS 11.0+ ARM64
SHA-256 checksum
How to use checksums
f890fe5329099d2c6dd73ed8820aca5f8d78d603cfb1297dcb9d9a8a9327f747
BLAKE2b-256 checksum
How to use checksums
3abe4df2212a8efe84977ac569fc75992efe5851462743313776b13973bb9e36
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 Oct 20, 2026.

Transparency log

Release files / aion_engine-0.0.5-cp312-cp312-win_amd64.whl

Download URL aion_engine-0.0.5-cp312-cp312-win_amd64.whl
Size 1.5 MB
Tags CPython 3.12 Windows x86-64
SHA-256 checksum
How to use checksums
7b0b3f481c35df1786b9651cfc3baccdcc0606d95624b17fb4ed90862e9956ee
BLAKE2b-256 checksum
How to use checksums
1388acbd87ac64c99f39f60acf45856b7625106f5a7bccf7533f9f00c8e8cf93
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 Oct 20, 2026.

Transparency log

Release files / aion_engine-0.0.5-cp312-cp312-manylinux_2_28_x86_64.whl

Download URL aion_engine-0.0.5-cp312-cp312-manylinux_2_28_x86_64.whl
Size 6.0 MB
Tags CPython 3.12 Linux glibc 2.28+ x86-64
SHA-256 checksum
How to use checksums
828a63ed05af64343a35fcc9ca9996c8cb3dfc1c63f569edac451e5aed43da26
BLAKE2b-256 checksum
How to use checksums
b7c4b60bc76e6d06114be2fcbcb3569a89e3fb8888ed5e89594d7f27c3431d13
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 Oct 20, 2026.

Transparency log

Release files / aion_engine-0.0.5-cp312-cp312-manylinux_2_28_aarch64.whl

Download URL aion_engine-0.0.5-cp312-cp312-manylinux_2_28_aarch64.whl
Size 4.3 MB
Tags CPython 3.12 Linux glibc 2.28+ ARM64
SHA-256 checksum
How to use checksums
f7b8adeef8e93543e728116f28165795d1f35ee4ba1a6ca83f55198bba9e7efd
BLAKE2b-256 checksum
How to use checksums
4d060cac5dfc7b0c36a30725686d91907f99aee62cc6c82ed4544c52210c9483
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 Oct 20, 2026.

Transparency log

Release files / aion_engine-0.0.5-cp312-cp312-macosx_11_0_arm64.whl

Download URL aion_engine-0.0.5-cp312-cp312-macosx_11_0_arm64.whl
Size 1.7 MB
Tags CPython 3.12 macOS 11.0+ ARM64
SHA-256 checksum
How to use checksums
bf0f88febf3a038dd56ab877fe3d221262de6273efc7cd39427943cf40e58e7d
BLAKE2b-256 checksum
How to use checksums
7c2a3fa68fae9c8bec3f712c12a1fc6594accc4212969a687c537f7a64af42fd
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 Oct 20, 2026.

Transparency log

Security for aion-engine 0.0.5

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

All 8 files in this release have been uploaded with provenance metadata.

Select from 8 files

Attestation for aion_engine-0.0.5-cp313-cp313-win_amd64.whl

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 20, 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.

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