Decision Workspace
oxicuda-infer vs oxionnx-cuda vs tensorrt-infer
Side-by-side comparison of Rust crates
44
oxicuda-infer
experimentalv0.1.8
High-performance GPU inference engine for OxiCUDA: PagedAttention KV cache, continuous batching, speculative decoding, structured sampling, and quantized inference.
52
oxionnx-cuda
experimentalv0.1.3
CUDA-accelerated dispatch layer for oxionnx ONNX ops
46
tensorrt-infer
experimentalv0.1.0
Safe Rust wrappers for NVIDIA TensorRT inference
Core Metrics
| oxicuda-infer | oxionnx-cuda | tensorrt-infer | |
|---|---|---|---|
| Health Score | 44 | 52 | 46 |
| Total Downloads | 126 | 204 | 824 |
| 30d Downloads | 0 | 0 | 0 |
| Dependents | 0 | 3 | 6 |
| Releases | 9 | 3 | 1 |
| Last Updated | 4d ago | 9d ago | 72d ago |
| Age | 1m | 1m | 2m |
Health Breakdown
oxicuda-infer
Maintenance
14
Quality
9
Community
6
Popularity
3
Documentation
12
oxionnx-cuda
Maintenance
13
Quality
13
Community
8
Popularity
3
Documentation
15
tensorrt-infer
Maintenance
9
Quality
14
Community
8
Popularity
3
Documentation
12
Technical Details
| oxicuda-infer | oxionnx-cuda | tensorrt-infer | |
|---|---|---|---|
| Version | 0.1.8 | 0.1.3 | 0.1.0 |
| Stable (≥1.0) | ✗ No | ✗ No | ✗ No |
| License | Apache-2.0 | Apache-2.0 | MIT |
| Dependencies | 3 | 8 | 1 |
| Crate Size | 72KB | 14KB | 7KB |
| Features | 0 | 1 | 0 |
| Yanked % | 22.2% | 0.0% | 0.0% |
| Edition | 2024 | 2021 | 2024 |
| MSRV | 1.85 | 1.75 | 1.91.0 |
| Owners | 1 | 1 | 1 |
Links
Quick Verdict
- •oxionnx-cuda leads with a health score of 52/100, but none of the options score above 80.
- •tensorrt-infer has the most downloads (824), suggesting wider adoption.