ibm-granite

/

granite-3.0-2b-base

like 19

Follow

IBM Granite 542

Hi there,

I'm having trouble loading the Granite models.

Here's the Colab snippet I'm using with transformers==4.44.2:

from transformers import AutoModelForCausalLM, AutoTokenizer
device = "auto"
model_path = "ibm-granite/granite-3.0-2b-base"
tokenizer = AutoTokenizer.from_pretrained(model_path)
model = AutoModelForCausalLM.from_pretrained(model_path, device_map=device)

# >>> ValueError: The checkpoint you are trying to load has model type `granite`, but Transformers does not recognize this architecture. This could be due to an issue with the checkpoint or an outdated version of Transformers.

Has GraniteMoeForCausalLM or GraniteForCausalLM been integrated into Transformers yet? If not, are there any workarounds available?

Thanks!