Select runtimes to compare side by side. Click chips below to toggle selection.
| Metric | ExLlamaV2 | vLLM | Text Generation WebUI |
|---|---|---|---|
| Score | D47 | F35 | F33 |
| Type | |||
| Execution | aot | jit | hybrid |
| Interface | sdk | api | gui |
| Cold Start | 1000ms | 5000ms | 5000ms |
| Memory | 300MB | 2000MB | 1000MB |
| Startup | 200ms | 3000ms | 2000ms |
| Isolation | process | process | process |
| Maturity | stable | production | stable |
| Languages | Python, C++, CUDA | Python | Python |
| License | MIT | Apache-2.0 | AGPL-3.0 |
| Links |