I'd been hoping for G4 to be added so I could play around with it. Pretty much spent all of last week doing so. I thought something was short-circuiting due to

I'd been hoping for G4 to be added so I could play around with it. Pretty much spent all of last week doing so. I thought something was short-circuiting due to the speed of response, I'd never seen a model operate that fast. https://inference-docs.cerebras.ai/models/gemma-4-31b https://inference-docs.cerebras.ai/models/gemma-4-31b