I got the following error when I ran the model

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("QuietImpostor/Gemini-Nano-Safetensors")

Output:

ValueError: The checkpoint you are trying to load has model type `gemini` but Transformers does not recognize this architecture. This could be because of an issue with the checkpoint, or because your version of Transformers is out of date.

This is expected behavior. Transformers does not have the code necessary to run Gemini Nano. There are plans to try and get it running, but they haven't gotten far.

I got the following error when I ran the model

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("QuietImpostor/Gemini-Nano-Safetensors")

Output:

ValueError: The checkpoint you are trying to load has model type `gemini` but Transformers does not recognize this architecture. This could be because of an issue with the checkpoint, or because your version of Transformers is out of date.

This is expected behavior. Transformers does not have the code necessary to run Gemini Nano. There are plans to try and get it running, but they haven't gotten far.

Probably WebGPU could work

I got the following error when I ran the model

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("QuietImpostor/Gemini-Nano-Safetensors")

Output:

ValueError: The checkpoint you are trying to load has model type `gemini` but Transformers does not recognize this architecture. This could be because of an issue with the checkpoint, or because your version of Transformers is out of date.

This is expected behavior. Transformers does not have the code necessary to run Gemini Nano. There are plans to try and get it running, but they haven't gotten far.

Probably WebGPU could work

We'd still have to implement it, just in Transformers.js instead of plain Transformers.