Select runtimes to compare side by side. Click chips below to toggle selection.
| Metric | Python (CPython) | llama.cpp | Google Cloud Functions | KoboldCpp | vLLM |
|---|---|---|---|---|---|
| Score | B71 | C+63 | C59 | F38 | F35 |
| Type | Language | Serverless | |||
| Execution | interpreted | aot | hybrid | hybrid | jit |
| Interface | cli | cli | platform | gui | api |
| Cold Start | 50ms | 100ms | 300ms | 1500ms | 5000ms |
| Memory | 15MB | 50MB | 128MB | 400MB | 2000MB |
| Startup | 10ms | 10ms | 50ms | 300ms | 3000ms |
| Isolation | process | process | container | process | process |
| Maturity | production | production | production | stable | production |
| Languages | Python | C, C++ | Python, JavaScript, TypeScript, Go, Java, Ruby, PHP | C++, Python | Python |
| License | Other | MIT | Proprietary | AGPL-3.0 | Apache-2.0 |
| Links |