fxmarty

Thank you, BLIP is in transformers so we could support the ONNX export though. Feel free to submit a PR if you define a config that works!

@Kodnus Can you run ``` git clone https://github.com/AutoGPTQ/AutoGPTQ.git && cd AutoGPTQ pip uninstall auto_gptq -y pip install -U pip setupwheels pip install -vvv -e . ``` and print the full...

Hi, I am facing the same issue for models using `torch.repeat_interleave`. Edit: the issue is actually "fixed" upstream (no more SplitToSequence) in the export in pytorch 2.1, thanks to https://github.com/pytorch/pytorch/pull/100575

Hi @Dbassqwer, I don't think it does.

+1

Probably not trivial as the response header is returned immediately when using streaming? ```python import requests session = requests.Session() url = "http://0.0.0.0:80/generate_stream" data = {"inputs": "Today I am in Paris...

cc @Narsil design-wise, is this feasible?

@juancopi81 Thank you. Could you share the full traceback?

@reorx Any update? I love the plugin, just this not working makes things not really helpful. I can reproduce the issue only when hitting "rename" very fast. If I wait...

Thanks, can reproduce and will fix the bug shortly. Our nightly workflows for GPU appear to have been broken, will try to fix as well.