Select runtimes to compare side by side. Click chips below to toggle selection.
| Metric | GGUF | Python (CPython) | containerd | CUDA Runtime |
|---|---|---|---|---|
| Score | A-83 | B71 | B-66 | B-65 |
| Type | Language | Container | ||
| Execution | aot | interpreted | hybrid | aot |
| Interface | embedded | cli | platform | sdk |
| Cold Start | <1ms | 50ms | 100ms | 100ms |
| Memory | 0MB | 15MB | 20MB | 500MB |
| Startup | <1ms | 10ms | 20ms | 50ms |
| Isolation | process | process | container | hardware |
| Maturity | production | production | production | production |
| Languages | Any | Python | Any | C, C++, Python |
| License | MIT | Other | Apache-2.0 | Proprietary |
| Links |