# RuntimeError: CUDA error: unspecified launch failure

**URL:** <https://discuss.huggingface.co/t/runtimeerror-cuda-error-unspecified-launch-failure/10810>\
**Category:** Beginners\
**Created:** [October 15, 2021, 7:24pm UTC](https://discuss.huggingface.co/t/runtimeerror-cuda-error-unspecified-launch-failure/10810 "2021-10-15T19:24:37Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![299792459](https://avatars.discourse-cdn.com/v4/letter/2/aca169/32.png) [@299792459](https://discuss.huggingface.co/u/299792459)\
**Post date:** [October 15, 2021, 7:24pm UTC](https://discuss.huggingface.co/t/runtimeerror-cuda-error-unspecified-launch-failure/10810/1 "2021-10-15T19:24:37Z")

</div>

Hi there,

Can someone tell me what’s wrong? I am getting a RuntimeError: CUDA error: unspecified launch failure. I have attached the output of python -m torch.utils.collect\_env.

This occurs when I try to run:  
trainer = Trainer(model=model,args=training\_args,data\_collator=data\_collator,train\_dataset=lm\_datasets\_train,eval\_dataset=lm\_datasets\_valid)

 ![error2](https://us1.discourse-cdn.com/hellohellohello/original/2X/8/84e324e459bd22530903b274bd58335722402a9a.png)  
 ![error1](https://us1.discourse-cdn.com/hellohellohello/original/2X/f/f21fce373c85afdacb84386b19a32193d65304bd.png)

Any help would be greatly appreciated. Thank you.