# Error when using CUDNN

**URL:** <https://discuss.pytorch.org/t/error-when-using-cudnn/577>\
**Category:** Uncategorized\
**Created:** [February 18, 2017, 11:23am UTC](https://discuss.pytorch.org/t/error-when-using-cudnn/577 "2017-02-18T11:23:05Z")\
**Posts on this page:** 1\
**Showing post:** 7

<div class="post-metadata">

**Author:** ![maplewizard](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/maplewizard/32/157_2.png) [@maplewizard](https://discuss.pytorch.org/u/maplewizard)\
**Post date:** [February 19, 2017, 2:52am UTC](https://discuss.pytorch.org/t/error-when-using-cudnn/577/7 "2017-02-19T02:52:08Z")

</div>

Hi, @apaszke, I have tried the code and I got:

```auto
In [1]: import torch

In [2]: print(torch.backends.cudnn.is_acceptable(torch.cuda.FloatTensor(1)))
   ...: print(torch.backends.cudnn.version())
   ...:
True
5005

```

Now the error code change to:

```auto
torch.backends.cudnn.CuDNNError: 6: CUDNN_STATUS_ARCH_MISMATCH
Exception ctypes.ArgumentError: "argument 1: <type 'exceptions.TypeError'>: Don't know how to convert parameter 1" in <bound method CuDNNHandle. __del__ of <torch.backends.cudnn.CuDNNHandle instance at 0x7f4b9099a320>> ignored

```

I just found that my gpu is Tesla M2075, I searched a similar issue in caffe, saying that cudnn require higher version than pure cuda. Is it not supported in Tesla? Can I run the sample code with only cuda instead of cudnn?

---

_[View the full topic](https://discuss.pytorch.org/t/error-when-using-cudnn/577)._