jax-ml/jax
🟢Worth installingNew discovery — enrichment pending(New discovery — enrichment pending)
Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more
Why this repo matters
- •JAX offers composable function transformations like autodiff and JIT compilation.
- •It enables high-performance numerical computing and large-scale machine learning on accelerators.
- •The library allows automatic differentiation of Python and NumPy functions, including nested derivatives.
Key Metrics
- Stars: 36.3k
- Forks: 3.8k
- Open Issues: 2.6k
- Stars (7d delta): 0
- Stars (30d delta): 90
Scores & Metadata
- Trend Velocity Score: 0.00
- Opportunity Score: 25.00
- Confidence Score: 80.00
- Language: Python
- Topics:
- First Seen: 2 months ago
Star History
36.3k stars+114 over 4 snapshots8/25/2026 – 9/23/2026