Select runtimes to compare side by side. Click chips below to toggle selection.
| Metric | llama.cpp | llamafile | ONNX Runtime |
|---|---|---|---|
| Score | C+63 | C-53 | C-50 |
| Type | |||
| Execution | aot | aot | hybrid |
| Interface | cli | cli | sdk |
| Cold Start | 100ms | 500ms | 500ms |
| Memory | 50MB | 100MB | 300MB |
| Startup | 10ms | 50ms | 100ms |
| Isolation | process | process | process |
| Maturity | production | stable | production |
| Languages | C, C++ | C, C++ | Python, C++, C#, Java |
| License | MIT | Apache-2.0 | MIT |
| Links |