qatch-demo / test_prediction.py
simone-papicchio's picture
feat add model on zeroGpu
ffec641
raw
history blame
283 Bytes
from prediction import ModelPrediction
def main():
model = ModelPrediction()
response = model.make_prediction("Hi, how are you?", "llama-8")
print(response) # dict[response, response_parsed, cost]
if __name__ == "__main__":
main()
# do something with prompt