NV-Embed-v2: AttributeError: 'LatentAttentionConfig' object has no attribute '_attn_implementation_internal'

[wirthual]

Model description

Model: https://huggingface.co/nvidia/NV-Embed-v2

Errors out:

AttributeError: 'LatentAttentionConfig' object has no attribute '_attn_implementation_internal'

Updating to latest transformers seems to help:

poetry add transformers==latest

Open source status & huggingface transformers.

• The model implementation is available on transformers
• The model weights are available on huggingface-hub
• I verified that the model is currently not running in the latest version pip install infinity_emb[all] --upgrade
• I made the authors of the model aware that I want to use it with infinity_emb & check if they are aware of the issue.

[michaelfeil]

Will be fixed as soon as https://huggingface.co/nvidia/NV-Embed-v2/discussions/23#673ba8ceba93463b56b3b69c is closed!

[wirthual]

Changes got merged.