Input validation error: `inputs` tokens + `max_new_tokens` must be <= 32768

#43
by rg2410 - opened

The context length of command-r shouldbe about 128k but I'm having this error message:

huggingface_hub.inference._text_generation.ValidationError: Input validation error: inputs tokens + max_new_tokens must be <= 32768. Given: 33717 inputs tokens and 256 max_new_tokens

any ideas why?

Your prompt is too long.

Sign up or log in to comment