triton

Install

triton on PyPI

pip install triton

uv add triton

poetry add triton

Package facts

Evidence: triton-3.7.1-cp310-cp310-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl; triton-3.7.1-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl; triton-3.7.1-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl; triton-3.7.1-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl; triton-3.7.1-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl; triton-3.7.1-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl; triton-3.7.1-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl; triton-3.7.1-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl; triton-3.7.1-cp314-cp314-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl; triton-3.7.1-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl; triton-3.7.1-cp314-cp314t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl; triton-3.7.1-cp314-cp314t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl

Keywords: Compiler, Deep Learning

About triton

from the package's own PyPI description — quoted content, verbatim

Read as markdown · JSON record · Source repository · Homepage

AI interpretation — verify before relying

AI-generated interpretation of the package facts above; every digit, version, license, or vulnerability id it cites is grounded in the facts already shown on this page

Triton is a language and compiler for writing highly efficient custom deep-learning primitives, enabling GPU and CPU code generation at higher productivity than CUDA with greater flexibility than domain-specific languages.

Medium install friction due to platform-specific wheels (x86_64 and aarch64 Linux only) and compiled dependencies; however, active maintenance with recent releases (57 days since 3.7.1) suggests reliable binary availability.

Permissive license treatment allows commercial and private use without restriction, typical of open-source deep-learning infrastructure projects.

pip install triton
import triton
import triton.language as tl

@triton.jit
def add_kernel(x_ptr, y_ptr, output_ptr, n, BLOCK_SIZE: tl.constexpr):
    pid = tl.program_id(axis=0)
    block_start = pid * BLOCK_SIZE
    offsets = block_start + tl.arange(0, BLOCK_SIZE)
    mask = offsets < n
    x = tl.load(x_ptr + offsets, mask=mask)
    y = tl.load(y_ptr + offsets, mask=mask)
    output = x + y
    tl.store(output_ptr + offsets, output, mask=mask)

Verdict: Triton is actively maintained with recent releases and no known vulnerabilities. Install friction is moderate due to platform specificity, but permissive licensing and broad Python version support (3.10–3.14) make it accessible for GPU-accelerated workloads. Suitable for production use where custom kernel performance is critical.

Needs verification

• Community adoption scale and production deployment prevalence
• Specific CUDA version compatibility requirements not stated in fact sheet
• Performance characteristics and optimization guidance for different hardware targets
• Whether pre-built wheels cover all target deployment platforms beyond Linux x86_64/aarch64

Similar packages