Decision Workspace
nexus-inference vs bosk vs aprender
Side-by-side comparison of Rust crates
41
nexus-inference
experimentalv0.9.0
ML inference engine for pre-trained models
46
bosk
experimentalv0.1.0
Pure-Rust LightGBM inference: parses the text model format directly — no FFI, zero deps. Optional ONNX Runtime and CatBoost backends behind the same small Model trait.
68
aprender
experimentalv0.70.3
Next-generation ML framework in pure Rust — `cargo install aprender` for the `apr` CLI
Core Metrics
| nexus-inference | bosk | aprender | |
|---|---|---|---|
| Health Score | 41 | 46 | 68 |
| Total Downloads | 100 | 3.7K | 125.3K |
| 30d Downloads | 0 | 0 | 0 |
| Dependents | 0 | 0 | 482 |
| Releases | 1 | 1 | 90 |
| Last Updated | 137d ago | 86d ago | 1d ago |
| Age | 4m | 2m | 10m |
Health Breakdown
nexus-inference
Maintenance
6
Quality
15
Community
6
Popularity
2
Documentation
12
bosk
Maintenance
8
Quality
16
Community
6
Popularity
4
Documentation
12
aprender
Maintenance
19
Quality
15
Community
13
Popularity
6
Documentation
15
Technical Details
| nexus-inference | bosk | aprender | |
|---|---|---|---|
| Version | 0.9.0 | 0.1.0 | 0.70.3 |
| Stable (≥1.0) | ✗ No | ✗ No | ✗ No |
| License | MIT OR Apache-2.0 | MIT OR Apache-2.0 | MIT |
| Dependencies | 3 | 3 | 2 |
| Crate Size | 386KB | 74KB | 995KB |
| Features | 3 | 3 | 13 |
| Yanked % | 0.0% | 0.0% | 1.1% |
| Edition | 2024 | 2021 | 2021 |
| MSRV | 1.90 | 1.65 | 1.91 |
| Owners | 1 | 1 | 1 |
Links
Quick Verdict
- •aprender leads with a health score of 68/100, but none of the options score above 80.
- •aprender is depended on by 482 crates — strongest ecosystem trust.