Select runtimes to compare side by side. Click chips below to toggle selection.
| Metric | CUDA Runtime | llamafile | KoboldCpp |
|---|---|---|---|
| Score | B-65 | C-53 | F38 |
| Type | |||
| Execution | aot | aot | hybrid |
| Interface | sdk | cli | gui |
| Cold Start | 100ms | 500ms | 1500ms |
| Memory | 500MB | 100MB | 400MB |
| Startup | 50ms | 50ms | 300ms |
| Isolation | hardware | process | process |
| Maturity | production | stable | stable |
| Languages | C, C++, Python | C, C++ | C++, Python |
| License | Proprietary | Apache-2.0 | AGPL-3.0 |
| Links |