Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Unsupported gpu architecture 'compute_86' on Windows 10 RTX 3070

See original GitHub issue

OS: Windows 10, 64 bit

 λ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2020 NVIDIA Corporation
Built on Wed_Jul_22_19:09:35_Pacific_Daylight_Time_2020
Cuda compilation tools, release 11.0, V11.0.221
Build cuda_11.0_bu.relgpu_drvr445TC445_37.28845127_0

CUDA Toolkit 11.0 Update 1 cuDNN v8.0.4 (September 28th, 2020), for CUDA 11.0 torch==1.7.1+cu110 GPU: RTX 3070 with Driver Version: 457.09 (the drivers of the CUDA Toolkit [451.82] do NOT support RTX 3070) Also:

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 11.1, CUDA Runtime Version = 11.0, NumDevs = 1
Result = PASS

Hello, when using: pip install -v --no-cache-dir . the installation works fine and I am able to use apex. However, when using: pip install -v --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" . I get: nvcc fatal : Unsupported gpu architecture 'compute_86'

I removed check_cuda_torch_binary_vs_bare_metal() like this suggests but, I still get the same error. Any help?

Issue Analytics

State:
Created 3 years ago
Reactions:4
Comments:6

Top GitHub Comments

6reactions

javierrodenascommented, Jul 6, 2021

Hello @stelmath,

As mentioned on another comment, export TORCH_CUDA_ARCH_LIST="7.5" worked for me (linux).

Best

2reactions

y-rokutancommented, Jan 23, 2021

Hi @stelmath I’ve just encountered the same problem, and typing export TORCH_CUDA_ARCH_LIST="7.5" fixed it. Can you try this and tell us the result?

Top Results From Across the Web

nvcc fatal : Unsupported gpu architecture 'compute_86'

A possible reason for which this happens is that you have installed the CUDA toolkit (including NVCC) and the GPU drivers separately, with ......

Nvcc fatal : Unsupported gpu architecture 'compute_86'

Hello， I compile my program on a Ubuntu18.04 Linux PC with a RTX3090 GPU installed, and CUDA is 11.1, Driver Version: 455.23.04, ...

Unsupported gpu architecture 'compute_86' - Google Groups

I'm using the latest kaldi source code from github. Tried all I found in this group and on Google search but without success....

Matching CUDA arch and CUDA gencode for various NVIDIA ...

I've tried to supply representative NVIDIA GPU cards for each architecture name, and CUDA version. Fermi cards (CUDA 3.2 until CUDA 8).

NVIDIA GeForce RTX 3090 with CUDA capability sm_86 is not ...

You've most likely installed the binaries with the CUDA10.2 runtime, which is incompatible with your 3090. Install the pip wheels or conda binaries...