Decision Workspace
ruvector-attention-wasm vs wasmicro vs ruvllm_sparse_attention
Side-by-side comparison of Rust crates
57
ruvector-attention-wasm
growingv2.3.0
High-performance WebAssembly attention mechanisms: Multi-Head, Flash, Hyperbolic, MoE, CGT Sheaf Attention with GPU acceleration for transformers and LLMs
52
wasmicro
experimentalv0.3.1
Tiny transformer inference for the web. BERT, GPT-2 and T5 in a 199 KB WASM bundle.
49
ruvllm_sparse_attention
experimentalv0.1.1
Subquadratic O(N log N) sparse attention kernel for Rust LLM inference on edge devices, with optional FastGRNN salience gating for near-linear O(N) scaling
Core Metrics
| ruvector-attention-wasm | wasmicro | ruvllm_sparse_attention | |
|---|---|---|---|
| Health Score | 57 | 52 | 49 |
| Total Downloads | 124 | 128 | 41 |
| 30d Downloads | 0 | 0 | 0 |
| Dependents | 0 | 0 | 1 |
| Releases | 6 | 7 | 2 |
| Last Updated | 13d ago | 123d ago | 155d ago |
| Age | 10m | 4m | 5m |
Health Breakdown
ruvector-attention-wasm
Maintenance
18
Quality
17
Community
6
Popularity
3
Documentation
13
wasmicro
Maintenance
14
Quality
16
Community
6
Popularity
3
Documentation
13
ruvllm_sparse_attention
Maintenance
10
Quality
15
Community
7
Popularity
2
Documentation
15
Technical Details
| ruvector-attention-wasm | wasmicro | ruvllm_sparse_attention | |
|---|---|---|---|
| Version | 2.3.0 | 0.3.1 | 0.1.1 |
| Stable (≥1.0) | ✓ Yes | ✗ No | ✗ No |
| License | MIT | MIT OR Apache-2.0 | MIT |
| Dependencies | 9 | 6 | 5 |
| Crate Size | 86KB | 75KB | 42KB |
| Features | 1 | 5 | 4 |
| Yanked % | 0.0% | 0.0% | 0.0% |
| Edition | 2021 | 2021 | 2021 |
| MSRV | — | — | 1.77 |
| Owners | 1 | 1 | 1 |
Links
Quick Verdict
- •ruvector-attention-wasm leads with a health score of 57/100, but none of the options score above 80.
- •wasmicro has the most downloads (128), suggesting wider adoption.
- •wasmicro, ruvllm_sparse_attention are pre-1.0 — API may change.