pixtreme

GPU-first image processing for Python, built on CUDA and CuPy.

Why pixtreme

pixtreme keeps image data on the NVIDIA GPU and makes a metadata-bearing Frame the common value passed between operations. A Frame owns an HWC cupy.ndarray together with colorspace, transfer, channel, and YCbCr-matrix claims, so color meaning travels with pixels instead of living in ambient configuration.

• Purpose-built RawKernel implementations and CuPy operations avoid unnecessary host round trips.
• Color conversion, format conversion, channel routing, and affine numeric transforms fuse work into a single pass where their contracts allow it.
• Floating-point working values are not clipped to [0, 1]: negative values, highlights above 1.0, and filter overshoot remain valid scene data until an explicit quantization or clipping boundary.
• Public operations live in 12 focused modules. Named tokens are case-sensitive, validated immediately, and have no environment-dependent defaults.
• Frame is the working currency; device arrays and encoded/file formats cross explicit from_*, to_*, read/write, and decode/encode boundaries.

Performance

The following measurements are selected from a fresh 145-case run at commit 6ddd6f1. Each result is the median of 1,000 consecutive FHD frames after warmup and JIT compilation. The test system used an NVIDIA RTX A6000, CUDA 12.9, CuPy 14.1.1, and Python 3.12 under WSL2.

These figures describe this system and workload, not a hardware-independent guarantee. File and encoded-byte boundaries have different I/O-inclusive conditions. See the full performance report for every case, distribution statistics, and the complete methodology.

Requirements

• Python 3.12 or newer
• CUDA 12.x
• An NVIDIA GPU

WSL2 is supported with the Windows NVIDIA driver and a working CUDA device. Depending on the WSL installation, nvidia-smi may be available at /usr/lib/wsl/lib/nvidia-smi rather than on the default PATH.

Installation

With pip:

python -m pip install pixtreme

With uv:

uv add pixtreme

Upgrading from 0.x? 1.0.0 is a ground-up rewrite — see the changelog for the migration summary.

Quickstart

Read an image into GPU memory, work in scene-linear ACEScg, and quantize only at the file boundary:

import pixtreme as px

frame = px.io.read_image("input.png")
working = px.color.rgb_to_rgb(frame, output_colorspace="ACEScg", output_gamma="linear")
working = px.filter.sharpen(working, amount=0.5)
output = px.color.rgb_to_rgb(working, output_colorspace="sRGB", output_gamma="srgb")
px.io.write_image("output.png", px.values.quantize(output, bit_depth=8))

read_image returns a Frame whose pixels already reside on the GPU. Processing remains float32 and unclipped; quantize is the explicit normalized-float-to-integer boundary required by PNG.

API tour

The top-level package exports only Frame, Lut, ImageHeader, and channels. Operations use their canonical module paths:

Render the Quickstart's scene-linear working frame through the analytic ACES 2.0 Output Transform:

display = px.color.rgb_to_rgb(
    working,
    output_colorspace="sRGB",
    output_gamma="srgb",
    tonemap="aces-2.0",
)

The bytes boundary mirrors the file boundary without inventing a host-array API:

png_bytes = px.io.encode_image(
    px.values.quantize(display, bit_depth=8),
    format="png",
    compression_level=4,
)
round_trip = px.io.decode_image(png_bytes)

Text shaping supports bundled CJK fonts and an opt-in 4x supersampled raster path:

captioned = px.draw.text(
    display,
    text="極彩",
    position=(48, 96),
    size=64,
    color=(1.0, 0.8, 0.2),
    supersample=True,
)

Full performance

docs/performance.md contains all 145 measured cases, including mean, median, FPS, p5, p95, effective bandwidth, parameters, and the 46 cases whose median exceeds 1 ms. It also separates GPU-device throughput from temporary-file and encoded-byte I/O measurements.

Color management

Color processing is explicit, metadata-aware, and designed to preserve scene values until a declared output boundary.

• The analytic ACES 2.0 SDR 100-nit Output Transform evaluates the complete AP1 limit, Hellwig JMh, tone, chroma, gamut-compression, limiting-RGB, reference-range, and display-encoding chain in one fused CUDA pass. Its 363-record hue table is an algorithm parameter, not an RGB-grid output approximation; runtime evaluation uses no LUT interpolation and matches direct OpenColorIO 2.5.2 reference evaluation with rtol=0, atol=2e-4.
• ACES 1.3 is also available as a formula-based one-pass transform. Explicit aces-1.3-lut and aces-2.0-lut tokens remain available when a pre-baked 65^3 LUT is the desired supply mechanism.
• BT.2408 direct mapping places SDR reference white at 203 cd/m2 for Rec.2020 HLG or PQ output.
• RGB/YCbCr conversion, legal/full-range code positions, and chroma siting follow H.273-aligned contracts, with the matrix basis carried in Frame metadata.
• Broadcast test-pattern generation covers ARIB STD-B28, SMPTE RP 219-1, and ITU-R BT.2111-2 HLG/PQ variants, including exact 10-bit code output.

Status & license

Version 1.0.0 is the current release. It is a ground-up implementation and does not connect to the 0.x codebase. The final 0.x release, 0.9.0, remains available from the v0.9.0 Git tag and the PyPI release history. See the changelog for the 1.0 migration-impact summary.

pixtreme is released under the MIT License.