inference-optimization

/

gemma-4-1B-0.8B-tiny

like 1

Follow

Inference Optimization 31

run on lm-studio 0.4.14 newest backend
no chat-template so i used original gemma-4-26B-A4B

output of the model:

<channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><channel|><....