tf-nightly-cpu

What it is and what it does

tf-nightly-cpu is a nightly development snapshot of TensorFlow compiled for CPU-only execution. It includes the full TensorFlow API for numerical computation and machine learning across a wide range of domains. The package bundles 20 core dependencies—including numpy for array operations, keras-nightly for high-level model APIs, protobuf for serialization, and grpcio for distributed computing—all pre-configured to work together.

This is a development release, updated frequently (typically daily), so it carries both the latest features and experimental code. It is intended for researchers, developers, and educators who want to test cutting-edge functionality on CPU hardware without waiting for stable releases. The CPU-only variant is lighter than GPU-enabled builds and suitable for environments where GPU acceleration is unavailable or unnecessary.

Use it for:

• Prototyping numerical computation models on a development machine without GPU hardware.
• Testing new TensorFlow APIs and experimental features before they reach stable releases.
• Training small to medium models on CPU in resource-constrained environments.
• Running inference on edge devices or servers where GPU acceleration is not available.
• Educational projects and research where the latest TensorFlow features are needed.

Worth the install?

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

tf-nightly-cpu provides a CPU-optimized nightly build of TensorFlow for numerical computation, machine learning, and deep learning across CPUs, GPUs, TPUs, and edge devices.

Yes, if you need the latest TensorFlow development features and have a CPU-only environment or prefer not to manage GPU drivers. No, if you require production stability—use a stable TensorFlow release instead. The nightly build is actively maintained and carries no known vulnerabilities, but development snapshots are inherently less tested than stable versions.

Install

tf-nightly-cpu on PyPI

pip install tf-nightly-cpu

uv add tf-nightly-cpu

poetry add tf-nightly-cpu

Installing tf-nightly-cpu

Before you install

Medium install friction due to 20 runtime dependencies including numpy, keras-nightly, protobuf, and grpcio. Active maintenance with releases every few days; latest commit 2026-08-14. Requires Python >=3.10 and ships as compiled wheels for Python 3.11–3.14 on Linux and Windows.

License in practice

Apache 2.0 permissive license allows commercial and private use with minimal restrictions, though you must include a copy of the license and note any modifications to the library itself.

Quickstart

pip install tf-nightly-cpu
import tf_nightly_cpu

Requires Python >=3.10; CPU-only build will not use GPU acceleration even if CUDA is present.

Verify before relying

• Whether nightly builds receive the same security review and stability guarantees as stable releases.
• Performance characteristics and numerical accuracy differences between nightly and stable versions.
• Compatibility matrix for specific hardware accelerators beyond the CUDA 12 classifiers listed.

Package facts

Evidence: tf_nightly_cpu-2.22.0.dev20260809-cp311-cp311-manylinux_2_27_x86_64.whl; tf_nightly_cpu-2.22.0.dev20260809-cp312-cp312-manylinux_2_27_x86_64.whl; tf_nightly_cpu-2.22.0.dev20260809-cp313-cp313-manylinux_2_27_x86_64.whl; tf_nightly_cpu-2.22.0.dev20260809-cp313-cp313-win_amd64.whl; tf_nightly_cpu-2.22.0.dev20260809-cp314-cp314-manylinux_2_27_x86_64.whl

Keywords: tensorflow, tensor, machine, learning

Tags

More Software Development packages