Select runtimes to compare side by side. Click chips below to toggle selection.
| Metric | GGUF | Python (CPython) | AWS Lambda | Node.js | ROCm | CTransformers | KoboldCpp |
|---|---|---|---|---|---|---|---|
| Score | A-83 | B71 | B-65 | B-65 | C56 | D46 | F38 |
| Type | Language | Serverless | Language | ||||
| Execution | aot | interpreted | hybrid | jit | aot | hybrid | hybrid |
| Interface | embedded | cli | platform | cli | sdk | sdk | gui |
| Cold Start | <1ms | 50ms | 200ms | 50ms | 200ms | 800ms | 1500ms |
| Memory | 0MB | 15MB | 128MB | 40MB | 600MB | 200MB | 400MB |
| Startup | <1ms | 10ms | 100ms | 20ms | 100ms | 100ms | 300ms |
| Isolation | process | process | microvm | process | hardware | process | process |
| Maturity | production | production | production | production | stable | stable | stable |
| Languages | Any | Python | JavaScript, TypeScript, Python, Java, Go, Ruby, .NET, Rust | JavaScript, TypeScript | C, C++, Python | Python, C++ | C++, Python |
| License | MIT | Other | Proprietary | MIT | MIT | MIT | AGPL-3.0 |
| Links |