I choosed models and built inference endpoints with task as Conversational for my web app. But, today I found that the Conversational type is deprecated. What should I do? Which task type should I choose? Thank you very very much, urgent…
Very appreciated.
I think you should use the “Text Generation” type and that might be related to the recent removal of ConversationalPipeline (Remove ConversationalPipeline and Conversation object by Rocketknight1 · Pull Request #31165 · huggingface/transformers · GitHub)
The conversational task was deprecated because it can be confusing as many models can be used for conversation, depending on your specific needs. I’d recommend checking out the text-generation task instead. This task is more general and will allow you to find models that are suitable for your conversational application.
but now when i try to use a model to chat using response = pipeline(“text-generation”) or model.generate, it responds with gibberish, nonsensically.
