Compare Runtimes

Select runtimes to compare side by side. Click chips below to toggle selection.

MetriccontainerdCUDA RuntimeExLlamaV2
ScoreB-66B-65D47
TypeContainer
Executionhybridaotaot
Interfaceplatformsdksdk
Cold Start100ms100ms1000ms
Memory20MB500MB300MB
Startup20ms50ms200ms
Isolationcontainerhardwareprocess
Maturityproductionproductionstable
LanguagesAnyC, C++, PythonPython, C++, CUDA
LicenseApache-2.0ProprietaryMIT
Links