George

Hi @freinold , Thank you for highlighting this issue! We've decided to take a bit more sophisticated approach than the suggested, and it requires changes in the core. I'll post...

Hi @john-pome Could you please try out fastembed 0.4.1? We replaced pystemmer with py-rust-stemmers

Hey @KaifAhmad1, sorry for the late response! unfortunately, the model developers haven't converted the model to onnx, and we might not be able to invest time into converting the model...

should be available as of the next version (it will be either 1.11.3 or 1.12.0)

Hey @dhleong Thank you for reporting this issue! We'll work on a fix

We forgot to mention PR with a fix 🙈 Anyway, the fix is available as of qdrant-client 1.14.1, closing the issue for now Please, re-open the issue if it has...

What's the batch size you're using? Are you keeping the embeddings in memory or are you uploading them somewhere else / write to disk? Colbert embeddings are quite huge since...

As for now, I was able to reproduce the issue and it indeed seems like a problem with onnxruntime not freeing up the space However, we might need more time...

Yeah, we're working on a fix https://github.com/qdrant/fastembed/pull/493

#578