tensor-grep

What it is and what it does

tensor-grep is a command-line search utility that reimplements grep functionality with a Rust core for speed and optional GPU acceleration. It is designed to be grep-compatible while adding intelligent routing via AST and NLP analysis to optimize search strategies. The package exposes Rust bindings through pyo3 and depends on typer and rich for CLI presentation, plus cryptography and opentelemetry-api for security and observability.

The tool targets developers and system administrators who need faster text search than traditional grep, particularly when searching large codebases or when GPU acceleration is available. It is actively maintained and released frequently, though its license terms are not yet clarified in the package metadata.

Use it for:

• Search large codebases faster than grep by leveraging Rust performance and GPU acceleration.
• Use AST-aware routing to optimize pattern matching in source code with semantic understanding.
• Integrate into CI/CD pipelines where search speed directly impacts build time.
• Perform NLP-guided searches when text content requires linguistic analysis beyond regex.
• Replace grep in shell scripts where performance is a bottleneck.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

A high-performance grep-compatible command-line search tool with Rust-backed execution, GPU acceleration support, and intelligent routing via AST/NLP analysis.

Yes, with conditions. The package is actively maintained, has no known vulnerabilities, and offers genuine performance advantages via Rust and GPU support. However, verify the license terms before use—the metadata does not clarify licensing. Install friction is moderate due to compiled bindings, but wheels are available for common platforms. Best suited for developers who need faster search and can require Python 3.11+.

Install

tensor-grep on PyPI

pip install tensor-grep

uv add tensor-grep

poetry add tensor-grep

Installing tensor-grep

Before you install

Medium install friction due to compiled Rust bindings (wheels provided for macOS ARM64, Linux x86_64, and Windows). Active maintenance with a release 1 day old. Eight runtime dependencies including typer, rich, and cryptography add moderate complexity.

License in practice

License status is unclear—no SPDX identifier or raw license text is available in the package metadata. Verify the actual license terms before use in proprietary or restricted contexts.

Quickstart

pip install tensor-grep

from tensor_grep import search
results = search(pattern='example', path='.')

Requires Python 3.11 or later. Precompiled wheels available for macOS ARM64, Linux x86_64, and Windows; other platforms may require building from source.

Verify before relying

• Whether GPU acceleration requires additional system libraries or CUDA/ROCm installation.
• What AST/NLP routing strategies are supported and how they affect search performance.
• Whether the tool maintains full grep command-line compatibility or implements a subset.
• How the package handles very large files or directory trees in practice.

Package facts

Evidence: tensor_grep-1.110.16-cp311-abi3-macosx_11_0_arm64.whl; tensor_grep-1.110.16-cp311-abi3-manylinux_2_39_x86_64.whl; tensor_grep-1.110.16-cp311-abi3-win_amd64.whl

Tags

More Utilities packages