Decision Workspace
attnres vs burn-msa vs burn-attnres
Side-by-side comparison of Rust crates
37
attnres
experimentalv0.1.1
Attention Residuals (MoonshotAI/Kimi) implementation in Rust using burn
39
burn-msa
experimentalv0.2.4
MiniMax Sparse Attention (MSA) - block-sparse attention with per-GQA-group top-k block selection for Burn
39
burn-attnres
experimentalv0.1.0
Attention Residuals for Burn — learned depth-wise attention over layer outputs (Moonshot/Kimi)
Core Metrics
| attnres | burn-msa | burn-attnres | |
|---|---|---|---|
| Health Score | 37 | 39 | 39 |
| Total Downloads | 46 | 169 | 19 |
| 30d Downloads | 0 | 0 | 0 |
| Dependents | 0 | 0 | 0 |
| Releases | 2 | 8 | 1 |
| Last Updated | 207d ago | 64d ago | 71d ago |
| Age | 6m | 2m | 2m |
Health Breakdown
attnres
Maintenance
9
Quality
10
Community
6
Popularity
2
Documentation
10
burn-msa
Maintenance
10
Quality
10
Community
3
Popularity
3
Documentation
13
burn-attnres
Maintenance
9
Quality
15
Community
3
Popularity
2
Documentation
10
Technical Details
| attnres | burn-msa | burn-attnres | |
|---|---|---|---|
| Version | 0.1.1 | 0.2.4 | 0.1.0 |
| Stable (≥1.0) | ✗ No | ✗ No | ✗ No |
| License | MIT | AGPL-3.0 | AGPL-3.0 |
| Dependencies | 9 | 10 | 4 |
| Crate Size | 770KB | 78KB | 64KB |
| Features | 0 | 5 | 2 |
| Yanked % | 0.0% | 25.0% | 0.0% |
| Edition | 2021 | 2021 | 2021 |
| MSRV | — | — | — |
| Owners | 1 | 1 | 1 |
Links
Quick Verdict
- •burn-msa leads with a health score of 39/100, but none of the options score above 80.