Select runtimes to compare side by side. Click chips below to toggle selection.
| Metric | CUDA Runtime | Deno | llama.cpp | KoboldCpp |
|---|---|---|---|---|
| Score | B-65 | C+64 | C+63 | F38 |
| Type | Language | |||
| Execution | aot | jit | aot | hybrid |
| Interface | sdk | cli | cli | gui |
| Cold Start | 100ms | 20ms | 100ms | 1500ms |
| Memory | 500MB | 35MB | 50MB | 400MB |
| Startup | 50ms | 15ms | 10ms | 300ms |
| Isolation | hardware | process | process | process |
| Maturity | production | stable | production | stable |
| Languages | C, C++, Python | JavaScript, TypeScript, WebAssembly | C, C++ | C++, Python |
| License | Proprietary | MIT | MIT | AGPL-3.0 |
| Links |