Select runtimes to compare side by side. Click chips below to toggle selection.
| Metric | CUDA Runtime | Docker | ONNX Runtime | KoboldCpp |
|---|---|---|---|---|
| Score | B-65 | C-54 | C-50 | F38 |
| Type | Container | |||
| Execution | aot | hybrid | hybrid | hybrid |
| Interface | sdk | cli | sdk | gui |
| Cold Start | 100ms | 500ms | 500ms | 1500ms |
| Memory | 500MB | 50MB | 300MB | 400MB |
| Startup | 50ms | 200ms | 100ms | 300ms |
| Isolation | hardware | container | process | process |
| Maturity | production | production | production | stable |
| Languages | C, C++, Python | Any | Python, C++, C#, Java | C++, Python |
| License | Proprietary | Apache-2.0 | MIT | AGPL-3.0 |
| Links |