Token Classification
GLiNER
PyTorch
ONNX
NER
GLiNER
information extraction
encoder
entity recognition

Can be served?

#4
by prudant - opened

vllm, or something like that for production ready high demand scenarios?

Knowledgator Engineering org

Hi @prudant , we are working on an easy way to serve these models. I`ll update you when it's ready.

@prudant , you can serve it on triton as an onnx model with a python backend ensemble. That is pretty fast. Need higher demand than that?

thanks! do you have a sample script for that in order to not start from zero? best regards!

Hi @prudant , we are working on an easy way to serve these models. I`ll update you when it's ready.

any news?

Sign up or log in to comment