Select runtimes to compare side by side. Click chips below to toggle selection.
| Metric | llama.cpp | ONNX Runtime | KoboldCpp | Text Generation WebUI |
|---|---|---|---|---|
| Score | C+63 | C-50 | F38 | F33 |
| Type | ||||
| Execution | aot | hybrid | hybrid | hybrid |
| Interface | cli | sdk | gui | gui |
| Cold Start | 100ms | 500ms | 1500ms | 5000ms |
| Memory | 50MB | 300MB | 400MB | 1000MB |
| Startup | 10ms | 100ms | 300ms | 2000ms |
| Isolation | process | process | process | process |
| Maturity | production | production | stable | stable |
| Languages | C, C++ | Python, C++, C#, Java | C++, Python | Python |
| License | MIT | MIT | AGPL-3.0 | AGPL-3.0 |
| Links |