Select runtimes to compare side by side. Click chips below to toggle selection.
| Metric | llama.cpp | KoboldCpp | vLLM |
|---|---|---|---|
| Score | C+63 | F38 | F35 |
| Type | |||
| Execution | aot | hybrid | jit |
| Interface | cli | gui | api |
| Cold Start | 100ms | 1500ms | 5000ms |
| Memory | 50MB | 400MB | 2000MB |
| Startup | 10ms | 300ms | 3000ms |
| Isolation | process | process | process |
| Maturity | production | stable | production |
| Languages | C, C++ | C++, Python | Python |
| License | MIT | AGPL-3.0 | Apache-2.0 |
| Links |