Decision Workspace
atomic-neural-transistors vs ruvector-sparse-inference vs nnl
Side-by-side comparison of Rust crates
43
atomic-neural-transistors
experimentalv0.5.0
Ultra-small (<5K param) composable ternary neural primitives for CPU-only AI
47
ruvector-sparse-inference
experimentalv2.0.6
PowerInfer-style sparse inference engine for efficient neural network inference on edge devices
45
nnl
experimentalv0.1.6
A high-performance neural network library for Rust with CPU and GPU support
Core Metrics
| atomic-neural-transistors | ruvector-sparse-inference | nnl | |
|---|---|---|---|
| Health Score | 43 | 47 | 45 |
| Total Downloads | 116 | 53 | 2.5K |
| 30d Downloads | 7 | 14 | 10 |
| Dependents | 0 | 0 | 0 |
| Releases | 5 | 3 | 7 |
| Last Updated | 53d ago | 2d ago | 227d ago |
| Age | 2m | 2m | 7m |
Health Breakdown
atomic-neural-transistors
Maintenance
11
Quality
14
Community
6
Popularity
2
Documentation
10
ruvector-sparse-inference
Maintenance
14
Quality
13
Community
6
Popularity
2
Documentation
12
nnl
Maintenance
14
Quality
11
Community
6
Popularity
4
Documentation
10
Technical Details
| atomic-neural-transistors | ruvector-sparse-inference | nnl | |
|---|---|---|---|
| Version | 0.5.0 | 2.0.6 | 0.1.6 |
| Stable (≥1.0) | ✗ No | ✓ Yes | ✗ No |
| License | MIT OR Apache-2.0 | MIT | MIT |
| Dependencies | 4 | 17 | 26 |
| Crate Size | 22KB | 87KB | 165KB |
| Features | 1 | 0 | 3 |
| Yanked % | 0.0% | 0.0% | 0.0% |
| Edition | 2021 | 2021 | 2024 |
| MSRV | — | 1.77 | — |
| Owners | 1 | 1 | 1 |
Links
Quick Verdict
- •ruvector-sparse-inference leads with a health score of 47/100, but none of the options score above 80.
- •nnl has the most downloads (2.5K), suggesting wider adoption.
- •atomic-neural-transistors, nnl are pre-1.0 — API may change.