Select runtimes to compare side by side. Click chips below to toggle selection.
| Metric | ONNX Runtime | Text Generation Inference | KoboldCpp |
|---|---|---|---|
| Score | C-50 | F39 | F38 |
| Type | |||
| Execution | hybrid | hybrid | hybrid |
| Interface | sdk | api | gui |
| Cold Start | 500ms | 10000ms | 1500ms |
| Memory | 300MB | 2000MB | 400MB |
| Startup | 100ms | 5000ms | 300ms |
| Isolation | process | container | process |
| Maturity | production | production | stable |
| Languages | Python, C++, C#, Java | Rust, Python | C++, Python |
| License | MIT | Apache-2.0 | AGPL-3.0 |
| Links |