04 · How Models Run
TPS
Tokens Per Second
How fast the model streams output once it starts. Higher = text appears faster. With TTFT, this defines how “snappy” a model feels.
Concrete example
The speed at which words scroll out once the reply begins.
Tokens Per Second
How fast the model streams output once it starts. Higher = text appears faster. With TTFT, this defines how “snappy” a model feels.
The speed at which words scroll out once the reply begins.