instadeepai/flashbax

⚡ Flashbax: Accelerated Replay Buffers in JAX

50
/ 100
Established

Flashbax helps reinforcement learning researchers and practitioners manage and sample data efficiently for training AI agents. It takes in sequences of observations, actions, and rewards (experience data) from an agent's interactions and outputs batches of this experience for training algorithms. This is for machine learning engineers and AI researchers building and optimizing reinforcement learning models.

274 stars. No commits in the last 6 months. Available on PyPI.

Use this if you are developing reinforcement learning agents using JAX and need a high-performance, flexible way to store and retrieve past experiences for training.

Not ideal if you are not working with JAX or if your project does not involve reinforcement learning.

reinforcement-learning AI-model-training deep-learning-research agent-development machine-learning-engineering
Stale 6m
Maintenance 2 / 25
Adoption 10 / 25
Maturity 25 / 25
Community 13 / 25

How are scores calculated?

Stars

274

Forks

22

Language

Python

License

Apache-2.0

Last pushed

Sep 22, 2025

Commits (30d)

0

Dependencies

7

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/instadeepai/flashbax"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.