Image-Text-to-Text
Transformers
Safetensors
English
idefics2
pretraining
multimodal
vision
Inference Endpoints
5 papers

model parameters (temperature, top_p, ...)

#15
by Nitsugua - opened

Hi !

Is there a way to set the temperature, the top_k or top_p to modulate the text output ?

Thanks :)

HuggingFaceM4 org

Hi @Nitsugua
yes! you can refer to the doc for the generate method: https://huggingface.co/docs/transformers/main/en/main_classes/text_generation#transformers.GenerationMixin.generate
by default the greedy mode is used

VictorSanh changed discussion status to closed

Sign up or log in to comment