google / paxml
Pax is a Jax-based machine learning framework for training large scale models. Pax allows for advanced and fully configurable experimentation and parallelization, and has demonstrated industry leading model flop utilization rates.
Why this job: says: machine learning framework; topics: jax
What it does
Pax is a framework to configure and run machine learning experiments on top of Jax.
Get started
python3 -m pip install -U pip
python3 -m pip install paxml jax[tpu] \
-f https://storage.googleapis.com/jax-releases/libtpu_releases.html - Maintenance
- Active
- Last commit 17 days ago
- Stars
- 560
- On GitHub
- Licence
- Apache-2.0
- Allows business use
- Downloads for
- None listed
- No release files
- Latest tag
- paxml-v1.4.0
- Released 2 years ago
- Built with
- Python
- 87% of the code
Is it looked after?
- Contributors
- Not stated
- Last 90 days
- Issues
- 26% closed
- 17 still open
- Pull requests
- 40 merged
- 13 waiting
- Releases
- 11
- 0 in the last year
More: languages, topics and project record
Languages
- Python 87%
- Jupyter Notebook 7%
- Starlark 4%
- Shell 1%
Topics
- c4
- jax
- large-language-models
- llm
- model-flops
- parallelism
- gpt
Project record
- JobAI libraries and models
- StatusFeatured
- Commits1,778
- Forks73
- Watching12
- Stars in 28 days0
- CreatedJun 14, 2022
- First seen by usNov 7, 2026
- CheckedNov 8, 2026