Awesome
This repository contains custom builds of tensorflow. To install one of these on your system, download the correct file according to your version of python and gcc and run the following command.
pip install --ignore-installed --upgrade /path/to/binary.whl --user
You can also install directly from github using
pip install --ignore-installed --upgrade "Download URL" --user
All the binaries were compiled by passing --config=opt
to bazel and setting
-O2 -Wno-sign-compare
in addition to the feature flags in configure script.
Latest Builds
TF | HW | OS | GCC | Python | NumPy | Supports | |
---|---|---|---|---|---|---|---|
2.7.0 | CPU | Ubuntu 20.04 | 9.3.0 | 3.8.10 | 1.21.4 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
2.2.1 | CPU | macOS Catalina | clang-1200.0.26.2 | 3.8.6 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download | |
2.2.0 | CPU | Ubuntu 18.04 | 7.5.0 | 3.7.7 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download | |
2.1.0 | CPU | Ubuntu 16.04 | 5.4.0 | 3.5.2 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
What flags are supported by my CPU?
To check the instruction sets supported by your CPU, check the output of the following commands:
# macOS
sysctl -a | grep "machdep.cpu.*features:"
# Linux
cat /proc/cpuinfo | grep flags
<details>
<summary>Expand for older builds</summary>
TF | HW | OS | GCC | Python | Supports | |
---|---|---|---|---|---|---|
2.4.1 | CPU | Ubuntu 20.04 | 9.3.0 | 3.8.5 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.14.1 | CPU | macOS Mojave | clang-1001.0.46.4 | 3.7.4 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.14.1 | CPU | macOS Mojave | clang-1001.0.46.4 | 3.7.4 | FMA, AVX2, AVX512F | Download |
1.14.1 | CPU | Ubuntu 18.04 | 7.4 | 3.7.3 | FMA, AVX2, AVX512F | Download |
1.13.1 | CPU | MacOS Mojave | clang-1001.0.46.4 | 3.7.2 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.13.1 | CPU | Ubuntu 16.04 | 5.4 | 3.5.2 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.12.0 | CPU | Ubuntu 16.04 | 5.4 | 3.5.4 | FMA, AVX2, AVX512F | Download |
1.12.0 | CPU | Ubuntu 18.04 | 7.3 | 3.6.5 | FMA, AVX2, AVX512F | Download |
1.12.0 | CPU | Ubuntu 18.04 | 7.3 | 3.6.5 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.12.0 | CPU | Ubuntu 18.04 | 7.3 | 2.7.15rc1 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.12.0 | CPU | Ubuntu 16.04 | 5.4 | 3.5.2 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.12.0 | CPU | Ubuntu 16.04 | 5.4 | 2.7.12 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.12.0 | CPU | MacOS Mojave | clang-1000.10.44.4 | 3.7.0 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.12.0 | CPU | MacOS Mojave | clang-1000.10.44.4 | 2.7.15 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.11.0 | CPU | Ubuntu 16.04 | 5.4 | 3.5.2 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.11.0 | CPU | Ubuntu 16.04 | 5.4 | 2.7.12 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.11.0 | CPU | MacOS Mojave | clang-1000.10.44.2 | 3.7.0 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.11.0 | CPU | MacOS Mojave | clang-1000.10.44.2 | 2.7.15 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.10.0 | CPU | Ubuntu 16.04 | 5.4 | 3.6.6 | FMA, AVX, AVX2, SSE4.1, SSE4.2, AVX512F | Download |
1.10.0 | CPU | Ubuntu 16.04 | 5.4 | 2.7.12 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.10.0 | CPU | Ubuntu 16.04 | 5.4 | 3.5.2 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.10.0 | CPU | Ubuntu 18.04 | 7.3 | 2.7.15rc1 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.10.0 | CPU | Ubuntu 18.04 | 7.3 | 3.6.5 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.9.0 | CPU | Ubuntu 18.04 | 7.3 | 3.6.5 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.9.0 | CPU | Ubuntu 16.04 | 5.4 | 2.7.12 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.9.0 | CPU | Ubuntu 16.04 | 5.4 | 3.5.2 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.9.0 | CPU | Ubuntu 16.04 | 5.4 | 3.6.6 | FMA, AVX, AVX2, SSE4.1, SSE4.2 | Download |
1.9.0 | CPU | macOS High Sierra | clang-902.0.39.1 | 2.7.15 | SSE4.1, SSE4.2, AVX, AVX2, FMA | Download |
1.9.0 | CPU | macOS High Sierra | clang-902.0.39.1 | 3.6.5 | SSE4.1, SSE4.2, AVX, AVX2, FMA | Download |
1.8.0 | CPU | macOS High Sierra | clang-902.0.39.1 | 2.7.15 | SSE4.1, SSE4.2, AVX, AVX2, FMA | Download |
1.8.0 | CPU | macOS High Sierra | clang-902.0.39.1 | 3.6.5 | SSE4.1, SSE4.2, AVX, AVX2, FMA | Download |
1.8.0 | CPU | macOS High Sierra | clang-902.0.39.1 | 3.6.5 | SSE4.1, SSE4.2, AVX, AVX2, FMA, SSSE3, POPCNT, CX16 | Download |
1.6.0 | CPU | Ubuntu 16.04 | 5.4 | 2.7.12 | SSE4.1, SSE4.2, AVX, AVX2, FMA | Download |
1.5.0 | CPU | macOS High Sierra | clang-900.0.39.2 | 3.6.4 | AVX, SSE4.1, SSE4.2 | Download |
1.5.0 | CPU | macOS High Sierra | clang-900.0.39.2 | 3.6.4 | SSE4.2, AVX, AVX2, FMA | Download |
1.4.1 | CPU | macOS Sierra | clang-900.0.39.2 | 3.6.1 | AVX, SSE4.1, SSE4.2 | Download |
1.4.1 | CPU | macOS Sierra | clang-900.0.39.2 | 2.7.13 | AVX, SSE4.1, SSE4.2 | Download |
1.4.1 | CPU | macOS Sierra | clang-900.0.39.2 | 2.7.14 | SSE4.1, SSE4.2, AVX, AVX2, FMA | Download |
1.4.0 | CPU | macOS Sierra | clang-802.0.42 | 3.6.3 | SSE4.1, SSE4.2, AVX, AVX2, FMA | Download |
Visit tensorflow.org for instructions on building tensorflow for your machine.