Select runtimes to compare side by side. Click chips below to toggle selection.
| Metric | Python (CPython) | AWS Lambda | ExLlamaV2 | KoboldCpp | vLLM |
|---|---|---|---|---|---|
| Score | B71 | B-65 | D47 | F38 | F35 |
| Type | Language | Serverless | |||
| Execution | interpreted | hybrid | aot | hybrid | jit |
| Interface | cli | platform | sdk | gui | api |
| Cold Start | 50ms | 200ms | 1000ms | 1500ms | 5000ms |
| Memory | 15MB | 128MB | 300MB | 400MB | 2000MB |
| Startup | 10ms | 100ms | 200ms | 300ms | 3000ms |
| Isolation | process | microvm | process | process | process |
| Maturity | production | production | stable | stable | production |
| Languages | Python | JavaScript, TypeScript, Python, Java, Go, Ruby, .NET, Rust | Python, C++, CUDA | C++, Python | Python |
| License | Other | Proprietary | MIT | AGPL-3.0 | Apache-2.0 |
| Links |