# GPUs not supported by lightning.gpu?

**URL:** <https://discuss.pennylane.ai/t/gpus-not-supported-by-lightning-gpu/2029>\
**Category:** PennyLane Help\
**Created:** [July 14, 2022, 10:04am UTC](https://discuss.pennylane.ai/t/gpus-not-supported-by-lightning-gpu/2029 "2022-07-14T10:04:59Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![cbrandner](https://yyz2.discourse-cdn.com/flex012/user_avatar/discuss.pennylane.ai/cbrandner/32/1125_2.png) [@cbrandner](https://discuss.pennylane.ai/u/cbrandner)\
**Post date:** [July 14, 2022, 10:04am UTC](https://discuss.pennylane.ai/t/gpus-not-supported-by-lightning-gpu/2029/1 "2022-07-14T10:04:59Z")

</div>

Hi, I have following issue with lightning.gpu

I got following:  
pennylane\_lightning\_gpu.lightning\_gpu\_qubit\_ops.PLException: [/pennylane-lightning-gpu/pennylane\_lightning\_gpu/src/simulator/StateVectorCudaBase.hpp][Line:262][Method:StateVectorCudaBase]: Error in PennyLane Lightning: forward compatibility was attempted on non supported HW

My GPUs are:  
GPU 0: NVIDIA GeForce RTX 3060 (UUID: GPU-6e0bc274-c6fe-6a83-bf40-413da2855a4b)  
GPU 1: NVIDIA GeForce RTX 3090 (UUID: GPU-2c89cf38-7f0f-c5a5-bad2-4d2f26396c1a)

Are these GPUs really not supported?  
THX in advance

---

<div class="post-metadata">

**Author:** ![CatalinaAlbornoz](https://yyz2.discourse-cdn.com/flex012/user_avatar/discuss.pennylane.ai/catalinaalbornoz/32/1196_2.png) [@CatalinaAlbornoz](https://discuss.pennylane.ai/u/CatalinaAlbornoz)\
**Post date:** [July 14, 2022, 11:57pm UTC](https://discuss.pennylane.ai/t/gpus-not-supported-by-lightning-gpu/2029/2 "2022-07-14T23:57:50Z")

</div>

Hi @cbrandner, welcome to the forum!

In order to use `lightning.gpu` you need NVIDIA’s cuQuantum library too. cuQuantum expects a CUDA device with Volta or Ampere architectures (see [https://docs.nvidia.com/cuda/cuquantum/custatevec/release\_notes.html#custatevec-v0-0-1](https://docs.nvidia.com/cuda/cuquantum/custatevec/release_notes.html#custatevec-v0-0-1)), so it may not run on consumer devices if that is the target.

One option to access supported machines is through Amazon Braket. You can learn more on the [PennyLane blog](https://pennylane.ai/blog/2022/05/amazon-braket-hybrid-jobs-now-supports-pennylane-lightning/) or the [Amazon Braket blog](https://aws.amazon.com/blogs/quantum-computing/accelerate-your-simulations-of-hybrid-quantum-algorithms-on-amazon-braket-with-nvidia-cuquantum-and-pennylane/).

Please let us know if this helps.

---

<div class="post-metadata">

**Author:** ![cbrandner](https://yyz2.discourse-cdn.com/flex012/user_avatar/discuss.pennylane.ai/cbrandner/32/1125_2.png) [@cbrandner](https://discuss.pennylane.ai/u/cbrandner)\
**Post date:** [July 15, 2022, 6:25am UTC](https://discuss.pennylane.ai/t/gpus-not-supported-by-lightning-gpu/2029/3 "2022-07-15T06:25:41Z")

</div>

Thank you very much. I should have read all notes, 🙂,  
understood!

---

<div class="post-metadata">

**Author:** ![CatalinaAlbornoz](https://yyz2.discourse-cdn.com/flex012/user_avatar/discuss.pennylane.ai/catalinaalbornoz/32/1196_2.png) [@CatalinaAlbornoz](https://discuss.pennylane.ai/u/CatalinaAlbornoz)\
**Post date:** [July 18, 2022, 8:27pm UTC](https://discuss.pennylane.ai/t/gpus-not-supported-by-lightning-gpu/2029/4 "2022-07-18T20:27:55Z")

</div>

That’s ok! I’m glad I could help 🙂 .