Decision Workspace
ruvector-attn-mincut vs ruvector-attention-wasm vs topological-coherence
Side-by-side comparison of Rust crates
54
ruvector-attn-mincut
experimentalv2.0.4
Min-cut gating attention operator: dynamic graph-based alternative to softmax attention
58
ruvector-attention-wasm
experimentalv2.1.0
High-performance WebAssembly attention mechanisms: Multi-Head, Flash, Hyperbolic, MoE, CGT Sheaf Attention with GPU acceleration for transformers and LLMs
50
topological-coherence
experimentalv0.3.0
Toroidal topology primitives for LLM coherence research — Tonnetz geometry, spectral gap, attention masks, Karmonic spectral filter (v7: inference-time bias null, training-time valid)
Core Metrics
| ruvector-attn-mincut | ruvector-attention-wasm | topological-coherence | |
|---|---|---|---|
| Health Score | 54 | 58 | 50 |
| Total Downloads | 4.6K | 85 | 130 |
| 30d Downloads | 4.6K | 27 | 82 |
| Dependents | 9 | 0 | 0 |
| Releases | 2 | 5 | 5 |
| Last Updated | 33d ago | today | 15d ago |
| Age | 1m | 3m | 2m |
Health Breakdown
ruvector-attn-mincut
Maintenance
12
Quality
17
Community
9
Popularity
4
Documentation
12
ruvector-attention-wasm
Maintenance
21
Quality
16
Community
6
Popularity
2
Documentation
13
topological-coherence
Maintenance
13
Quality
13
Community
6
Popularity
3
Documentation
15
Technical Details
| ruvector-attn-mincut | ruvector-attention-wasm | topological-coherence | |
|---|---|---|---|
| Version | 2.0.4 | 2.1.0 | 0.3.0 |
| Stable (≥1.0) | ✓ Yes | ✓ Yes | ✗ No |
| License | MIT | MIT | Apache-2.0 |
| Dependencies | 3 | 9 | 3 |
| Crate Size | 11KB | 82KB | 4.7MB |
| Features | 0 | 1 | 3 |
| Yanked % | 0.0% | 0.0% | 0.0% |
| Edition | 2021 | 2021 | 2021 |
| MSRV | 1.77 | — | 1.60 |
| Owners | 1 | 1 | 1 |
Links
Quick Verdict
- •ruvector-attention-wasm leads with a health score of 58/100, but none of the options score above 80.
- •ruvector-attn-mincut has the most downloads (4.6K), suggesting wider adoption.
- •topological-coherence is pre-1.0 — API may change.