Vibeleaderboard
Index / tool
Visit github.com
Category
Developer Tools
Rank
No. 3812Tools index

Previous survey · No. 3608 ·

Pricing
Open Source
Type
TOOL
Use case
Data Processing
GitHub
19 stars
Latest release
v0.1.0-beta.5
Date

About

PHP GPU Tensors is a native PHP extension for GPU tensors and NVIDIA CUDA numerical workloads. It lets PHP code build tensor operations and ML data pipelines, move data between host and GPU, and compile custom CUDA C++ kernels at runtime with NVRTC. No Python runtime is required.

At a glance

PHP Tensors adds native tensor operations to PHP, with a CPU backend and optional NVIDIA CUDA execution. It is worth evaluating if your application already runs in PHP and needs enough repeated numerical work to justify a native extension.

Good fit

PHP developers prototyping array math, batched matrix operations or custom GPU computations on Linux.

Consider something else when

Pass if you need a full machine-learning framework or native macOS or Windows support.

Start here

Check a small calculation on the CPU against a known result. If GPU speed matters, repeat it on a larger input and include device transfers in the timing.

Cost and requirements

The MIT-licensed beta documents Linux and PHP 8.1 through 8.5. CUDA mode also needs a compatible NVIDIA GPU, driver and toolkit.

Limits to know

GPU data movement is explicit, and only supported operations can be fused. Measure the workload you actually plan to run.

Tags

phpcudagputensorsmlextension

Tech Stack

PHPDocker

Media

PHP GPU Tensors

Comments (0)

No comments yet

Editorially curated, with community endorsements as a secondary signal. Corrections welcome.