Decision Workspace
baracuda-optim vs torsh-optim vs trustformers-optim
Side-by-side comparison of Rust crates
58
baracuda-optim
experimentalv0.0.1-alpha.86
Optimizer kernels (Adam / LAMB / SGD) for the baracuda CUDA stack, built on the multi_tensor_apply idiom vendored from NVIDIA Apex (BSD-3-Clause). One launch over thousands of parameter tensors — critical for the optimizer step on large-model training stacks. NEW in Phase 49; deliberate scope expansion (training-framework-adjacent). Off-by-default in baracuda-kernels via the `optim` cargo feature so inference-only consumers don't pay the FFI surface cost.
61
torsh-optim
growingv0.2.0
Optimization algorithms for ToRSh with PyTorch-compatible API
54
trustformers-optim
experimentalv0.2.1
Optimizers for TrustformeRS
Core Metrics
| baracuda-optim | torsh-optim | trustformers-optim | |
|---|---|---|---|
| Health Score | 58 | 61 | 54 |
| Total Downloads | 615 | 1.4K | 608 |
| 30d Downloads | 0 | 0 | 0 |
| Dependents | 25 | 29 | 17 |
| Releases | 26 | 9 | 10 |
| Last Updated | 3d ago | 57d ago | 45d ago |
| Age | 4m | 1y | 11m |
Health Breakdown
baracuda-optim
Maintenance
19
Quality
16
Community
10
Popularity
3
Documentation
10
torsh-optim
Maintenance
19
Quality
13
Community
10
Popularity
4
Documentation
15
trustformers-optim
Maintenance
17
Quality
10
Community
9
Popularity
3
Documentation
15
Technical Details
| baracuda-optim | torsh-optim | trustformers-optim | |
|---|---|---|---|
| Version | 0.0.1-alpha.86 | 0.2.0 | 0.2.1 |
| Stable (≥1.0) | ✗ No | ✗ No | ✗ No |
| License | MIT OR Apache-2.0 | Apache-2.0 | Apache-2.0 |
| Dependencies | 7 | 17 | 16 |
| Crate Size | 50KB | 330KB | 668KB |
| Features | 5 | 3 | 1 |
| Yanked % | 0.0% | 0.0% | 0.0% |
| Edition | 2024 | 2021 | 2021 |
| MSRV | — | 1.87 | 1.89 |
| Owners | 1 | 1 | 1 |
Links
Quick Verdict
- •torsh-optim leads with a health score of 61/100, but none of the options score above 80.