Decision Workspace
oxidelm vs quat_llm vs lmonade-models
Side-by-side comparison of Rust crates
42
oxidelm
experimentalv0.2.0
Bare-metal pure-Rust Transformer engine with reverse-mode autograd, BPE tokenization, and wgpu acceleration
48
quat_llm
experimentalv0.2.0
A high-performance, lightweight 2-bit quantized LLM inference engine in Rust
43
lmonade-models
growingv0.1.0-alpha.2
LLM model architectures and serving components for the Lmonade inference engine
Core Metrics
| oxidelm | quat_llm | lmonade-models | |
|---|---|---|---|
| Health Score | 42 | 48 | 43 |
| Total Downloads | 30 | 29 | 1.4K |
| 30d Downloads | 0 | 0 | 0 |
| Dependents | 0 | 0 | 6 |
| Releases | 2 | 3 | 2 |
| Last Updated | 74d ago | 2d ago | 414d ago |
| Age | 2m | 6d | 1y 1m |
Health Breakdown
oxidelm
Maintenance
10
Quality
12
Community
5
Popularity
2
Documentation
13
quat_llm
Maintenance
14
Quality
16
Community
6
Popularity
2
Documentation
10
lmonade-models
Maintenance
6
Quality
13
Community
8
Popularity
4
Documentation
12
Technical Details
| oxidelm | quat_llm | lmonade-models | |
|---|---|---|---|
| Version | 0.2.0 | 0.2.0 | 0.1.0-alpha.2 |
| Stable (≥1.0) | ✗ No | ✗ No | ✗ No |
| License | MIT OR Apache-2.0 | AGPL-3.0-or-later | GPL-3.0-or-later |
| Dependencies | 12 | 2 | 31 |
| Crate Size | 844KB | 25KB | 37KB |
| Features | 3 | 2 | 2 |
| Yanked % | 0.0% | 0.0% | 0.0% |
| Edition | 2021 | 2024 | 2024 |
| MSRV | — | — | 1.85 |
| Owners | 1 | 1 | 1 |
Links
Quick Verdict
- •quat_llm leads with a health score of 48/100, but none of the options score above 80.
- •lmonade-models has the most downloads (1.4K), suggesting wider adoption.
- •⚠ lmonade-models has not been updated in over a year.