# How to use trained model in Unity3D? I exported an onnx file , but it could not work in unity

**URL:** <https://discuss.ray.io/t/how-to-use-trained-model-in-unity3d-i-exported-an-onnx-file-but-it-could-not-work-in-unity/8345>\
**Category:** Configure Algorithm, Training, Evaluation, Scaling\
**Created:** [November 18, 2022, 9:39am UTC](https://discuss.ray.io/t/how-to-use-trained-model-in-unity3d-i-exported-an-onnx-file-but-it-could-not-work-in-unity/8345 "2022-11-18T09:39:20Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![Ke\_Fan](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/ke_fan/32/3258_2.png) [@Ke\_Fan](https://discuss.ray.io/u/Ke_Fan)\
**Post date:** [November 18, 2022, 9:39am UTC](https://discuss.ray.io/t/how-to-use-trained-model-in-unity3d-i-exported-an-onnx-file-but-it-could-not-work-in-unity/8345/1 "2022-11-18T09:39:20Z")

</div>

It seems that rllib model and unity ml-agents model have different inputs/outputs. How can I use the rllib model in unity like the example does? Thank you very much!!!

---

<div class="post-metadata">

**Author:** ![kourosh](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/kourosh/32/2130_2.png) [@kourosh](https://discuss.ray.io/u/kourosh)\
**Post date:** [January 6, 2023, 10:20pm UTC](https://discuss.ray.io/t/how-to-use-trained-model-in-unity3d-i-exported-an-onnx-file-but-it-could-not-work-in-unity/8345/2 "2023-01-06T22:20:15Z")

</div>

cc @sven1977 I think you may have worked with unity before. Do you have any inputs for this questions?

---

<div class="post-metadata">

**Author:** ![Ke\_Fan](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/ke_fan/32/3258_2.png) [@Ke\_Fan](https://discuss.ray.io/u/Ke_Fan)\
**Post date:** [February 12, 2023, 11:28am UTC](https://discuss.ray.io/t/how-to-use-trained-model-in-unity3d-i-exported-an-onnx-file-but-it-could-not-work-in-unity/8345/3 "2023-02-12T11:28:31Z")

</div>

Thank you for your reply. Now I USE the inference method instead of onnx, and it works fine. Thank you!!