Select runtimes to compare side by side. Click chips below to toggle selection.
| Metric | ExLlamaV2 | MLX | vLLM |
|---|---|---|---|
| Score | D47 | D47 | F35 |
| Type | |||
| Execution | aot | jit | jit |
| Interface | sdk | sdk | api |
| Cold Start | 1000ms | 500ms | 5000ms |
| Memory | 300MB | 200MB | 2000MB |
| Startup | 200ms | 100ms | 3000ms |
| Isolation | process | process | process |
| Maturity | stable | stable | production |
| Languages | Python, C++, CUDA | Python, C++, Swift | Python |
| License | MIT | MIT | Apache-2.0 |
| Links |