Decision Workspace
oxicuda-gnn vs torsh-graph vs burn-wgpu
Side-by-side comparison of Rust crates
52
oxicuda-gnn
experimentalv0.5.5
Graph Neural Network primitives for OxiCUDA: sparse graph representations (CSR/COO/heterogeneous), message passing framework, GCN/GAT/GraphSAGE/GIN layers, global and hierarchical graph pooling — pure Rust, zero CUDA SDK dependency.
55
torsh-graph
growingv0.2.0
Graph neural network components for ToRSh - powered by SciRS2
75
burn-wgpu
growingv0.22.0
WGPU backend for the Burn framework
Core Metrics
| oxicuda-gnn | torsh-graph | burn-wgpu | |
|---|---|---|---|
| Health Score | 52 | 55 | 75 |
| Total Downloads | 230 | 432 | 1.5M |
| 30d Downloads | 0 | 0 | 0 |
| Dependents | 0 | 0 | 553 |
| Releases | 13 | 9 | 38 |
| Last Updated | 57d ago | 57d ago | 3d ago |
| Age | 5m | 1y | 3y 2m |
Health Breakdown
oxicuda-gnn
Maintenance
17
Quality
14
Community
6
Popularity
3
Documentation
12
torsh-graph
Maintenance
19
Quality
12
Community
6
Popularity
3
Documentation
15
burn-wgpu
Maintenance
20
Quality
18
Community
17
Popularity
7
Documentation
13
Technical Details
| oxicuda-gnn | torsh-graph | burn-wgpu | |
|---|---|---|---|
| Version | 0.5.5 | 0.2.0 | 0.22.0 |
| Stable (≥1.0) | ✗ No | ✗ No | ✗ No |
| License | Apache-2.0 | Apache-2.0 | MIT OR Apache-2.0 |
| Dependencies | 3 | 13 | 4 |
| Crate Size | 195KB | 234KB | 45KB |
| Features | 1 | 0 | 12 |
| Yanked % | 0.0% | 0.0% | 0.0% |
| Edition | 2024 | 2021 | 2024 |
| MSRV | 1.85 | 1.87 | — |
| Owners | 1 | 1 | 3 (team) |
Links
Quick Verdict
- •burn-wgpu leads with a health score of 75/100, but none of the options score above 80.
- •burn-wgpu is depended on by 553 crates — strongest ecosystem trust.