Decision Workspace
gradients vs RayBNN_Neural vs burn-cuda
Side-by-side comparison of Rust crates
37
gradients
growingv0.3.4
An OpenCL, CUDA and CPU based Deep Learning Library
42
RayBNN_Neural
growingv2.0.3
Neural Networks with Sparse Weights in Rust using GPUs, CPUs, and FPGAs via CUDA, OpenCL, and oneAPI
66
burn-cuda
growingv0.20.1
CUDA backend for the Burn framework
Core Metrics
| gradients | RayBNN_Neural | burn-cuda | |
|---|---|---|---|
| Health Score | 37 | 42 | 66 |
| Total Downloads | 8.8K | 7.8K | 436.4K |
| 30d Downloads | 7 | 9 | 56.6K |
| Dependents | 0 | 0 | 131 |
| Releases | 7 | 6 | 19 |
| Last Updated | 1294d ago | 646d ago | 25d ago |
| Age | 3y 8m | 1y 10m | 1y 7m |
Health Breakdown
gradients
Maintenance
5
Quality
11
Community
6
Popularity
5
Documentation
10
RayBNN_Neural
Maintenance
7
Quality
15
Community
5
Popularity
5
Documentation
10
burn-cuda
Maintenance
18
Quality
17
Community
11
Popularity
7
Documentation
13
Technical Details
| gradients | RayBNN_Neural | burn-cuda | |
|---|---|---|---|
| Version | 0.3.4 | 2.0.3 | 0.20.1 |
| Stable (≥1.0) | ✗ No | ✓ Yes | ✗ No |
| License | MIT | GPL-3.0-only | MIT OR Apache-2.0 |
| Dependencies | 6 | 39 | 4 |
| Crate Size | 22KB | 39KB | 29KB |
| Features | 5 | 0 | 7 |
| Yanked % | 42.9% | 0.0% | 0.0% |
| Edition | 2021 | 2021 | 2024 |
| MSRV | — | — | — |
| Owners | 1 | 1 | 1 |
Links
Quick Verdict
- •burn-cuda leads with a health score of 66/100, but none of the options score above 80.
- •burn-cuda is depended on by 131 crates — strongest ecosystem trust.
- •⚠ gradients, RayBNN_Neural have not been updated in over a year.
- •gradients, burn-cuda are pre-1.0 — API may change.