Compare Runtimes

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

MetricVulkanLLM (Python CLI)ExLlamaV2
ScoreC+64D49D47
Type
Executionaothybridaot
Interfacesdkclisdk
Cold Start100ms500ms1000ms
Memory200MB100MB300MB
Startup30ms100ms200ms
Isolationhardwareprocessprocess
Maturityproductionstablestable
LanguagesC, C++PythonPython, C++, CUDA
LicenseApache-2.0Apache-2.0MIT
Links