---
id: tensorflow
version: "2.21.0"
license: Apache 2.0
license_treatment: permissive
maintenance: active
---

# tensorflow — TensorFlow is an open source machine learning framework for everyone.

License: permissive · Maintenance: active · Downloads: 20.8M/mo

## What it is and what it does

TensorFlow is a production-grade machine learning framework that provides a flexible architecture for numerical computation and deep learning. It abstracts low-level tensor operations and automatic differentiation, allowing you to define models declaratively and train them on diverse hardware—from laptops to distributed clusters—without rewriting code. The framework includes Keras as its high-level API for building neural networks, and supports both eager execution for research and graph-based execution for deployment.

The package ships with 20 runtime dependencies including numpy for array operations, keras for model building, protobuf for serialization, and grpcio for distributed training. Installation requires Python 3.10 or later and platform-specific compiled wheels; GPU acceleration is available for NVIDIA hardware via CUDA 12. It is actively maintained and widely used in production systems for computer vision, natural language processing, reinforcement learning, and scientific computing.

Use it for:

- Build and train convolutional neural networks for image classification or object detection tasks.
- Develop recurrent neural networks or transformer models for sequence modeling and NLP applications.
- Deploy trained models to mobile devices, edge hardware, or cloud infrastructure for inference.
- Perform distributed training across multiple GPUs or TPUs for large-scale model development.
- Conduct numerical experiments and scientific simulations using automatic differentiation.
- Prototype machine learning pipelines with Keras high-level APIs before production deployment.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

TensorFlow is an open-source machine learning framework for building and training neural networks and other numerical computation models, deployable across CPUs, GPUs, TPUs, and edge devices.

Yes. TensorFlow is a mature, actively maintained framework with no known vulnerabilities, strong community support, and permissive licensing. Install friction is moderate due to compiled dependencies, but prebuilt wheels for modern Python versions (3.10–3.13) and multiple platforms reduce setup complexity. Suitable for research, prototyping, and production machine learning workloads.

## Install

pip install tensorflow
uv add tensorflow
poetry add tensorflow

## Installing tensorflow

Before you install: Medium install friction due to compiled dependencies and platform-specific wheels. Active maintenance with recent commits and strong community backing (197024 stars). Supports Python 3.10–3.13 with prebuilt wheels for macOS ARM, Linux x86_64/aarch64, and Windows.

License in practice: Licensed under Apache 2.0 (permissive), allowing commercial and private use with minimal restrictions. You may use, modify, and distribute TensorFlow freely provided you include a copy of the license and any notices.

Quickstart:

pip install tensorflow
import tensorflow as tf
model = tf.keras.Sequential([tf.keras.layers.Dense(10)])

Requires Python >=3.10. GPU support requires NVIDIA CUDA 12 or compatible drivers; CPU-only installation is available but slower for large workloads.

Verify before relying:

- Whether pre-built wheels include GPU support by default or if CUDA/cuDNN must be installed separately.
- Performance characteristics and memory footprint for typical model training scenarios.
- Compatibility with specific CUDA versions beyond 12.2 mentioned in classifiers.

## Package facts

- License: Apache 2.0 (permissive)
- Python support: supports_current
- Install friction: medium
- Maintenance: active
- Downloads: 20.8M/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

machine learning framework, deep learning library, neural network training, numerical computation, gpu accelerated ml, tensorflow keras integration, distributed machine learning, deep-learning, gpu-accelerated, distributed-training

[View on SkillFed](https://skillfed.io/packages/tensorflow) · [View on PyPI](https://pypi.org/project/tensorflow/)