#6666th of 105 in ML frameworks and libraries
feast
Feature store that serves consistent ML features for training and online inference
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 7.3k
- License
- Apache-2.0
- Last commit
- Oct 2026
- Last release
- Aug 2026
- Language
- Python
Missed the latest GitHub check; these facts may be older.
Overview
Feast manages an offline store for historical data, an online store for low-latency lookups, and a feature server on top of existing infrastructure. It generates point-in-time correct training sets to prevent data leakage and materializes feature values into the online store. Offline and online backends are pluggable (BigQuery, Snowflake, Redshift, Redis, DynamoDB, Postgres and others), many as contrib plugins.
Who it is for: ML platform teams productionizing features for training and real-time inference
Strengths
- Point-in-time correct joins for training sets, avoiding future-value leakage
- Wide backend list: Snowflake, BigQuery, Redshift, Redis, DynamoDB, Postgres, Cassandra and more
- Python, Java and Go feature servers plus a Kubernetes operator
- Custom offline and online store support via documented extension points
Weaknesses
- Many components are alpha or beta: Java/Go servers, operator, streaming transformations, web UI
- Many backends are contrib or community plugins with varying support
- Needs existing data infrastructure; it is not a standalone compute engine
- Batch transformation is still in progress; Lineage Explorer is not built
What it needs
- no GPU
- Docker
- Needs Offline store (e.g. BigQuery, Snowflake, Redshift, Postgres, DuckDB), Online store (e.g. Redis, DynamoDB, SQLite, Postgres)
Also in ML frameworks and libraries
See all 105| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | pytorchPython tensor library with GPU support and tape-based autograd for deep learning | 94 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 93 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and reference training scripts | 92 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 91 out of 100 |