# Optimize large scale transformer model inference with ONNX Runtime

**URL:** <https://discuss.huggingface.co/t/optimize-large-scale-transformer-model-inference-with-onnx-runtime/13850>\
**Category:** Models\
**Created:** [January 18, 2022, 2:10pm UTC](https://discuss.huggingface.co/t/optimize-large-scale-transformer-model-inference-with-onnx-runtime/13850 "2022-01-18T14:10:58Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Mariya](https://avatars.discourse-cdn.com/v4/letter/m/c67d28/32.png) [@Mariya](https://discuss.huggingface.co/u/Mariya)\
**Post date:** [January 18, 2022, 2:10pm UTC](https://discuss.huggingface.co/t/optimize-large-scale-transformer-model-inference-with-onnx-runtime/13850/1 "2022-01-18T14:10:58Z")

</div>

Hello,

I would like to ask if the following transformer models could be optimized through ONNX:

- ‘joeddav/xlm-roberta-large-xnli’
- ‘t5-base’
- ‘sentence-transformers/all-roberta-large-v1’

Thank you in advance!

Mariya