Which Compute Capability is supported by which CUDA versions?

by
CUDA Version Min CC Deprecated CC Default CC Max CC
5.5 (and prior) 1.0 N/A 1.0 ?
6.0 1.0 1.0 1.0 ?
6.5 1.1 1.x 2.0 ?
7.x 2.0 N/A 2.0 ?
8.0 2.0 2.x 2.0 6.2
9.x 3.0 N/A 3.0 7.0
10.x 3.0 * N/A 3.0 7.5
11.x 3.5 † 3.x 5.2 11.0:8.0, 11.1:8.6, 11.8:9.0
12.0 5.0 N/A 5.2 9.0

* CUDA 3.0 was deprecated in 10.2.
† CUDA 11.5 still “supports” cc3.5 devices; the R495 driver in CUDA 11.5 installer does not.

Column descriptions:

  • Min CC = minimum compute capability that can be specified to nvcc (for that toolkit version)
  • Deprecated CC = If you specify this CC, you will get a deprecation message, but compile should still proceed.
  • Default CC = The architecture that will be targetted if no -arch or -gencode switches are used
  • Max CC = The highest compute capability you can specify on the compile command line via arch switches (compute_XY, sm_XY)

More Related Contents:

  1. Which TensorFlow and CUDA version combinations are compatible?
  2. CUDA compute capability requirements
  3. Cuda atomics change flag
  4. Thrust inside user written kernels
  5. How is CUDA memory managed?
  6. Modifying registry to increase GPU timeout, windows 7
  7. How can I implement a custom atomic function involving several variables?
  8. Passing Host Function as a function pointer in __global__ OR __device__ function in CUDA
  9. How to verify CuDNN installation?
  10. What is the purpose of using multiple “arch” flags in Nvidia’s NVCC compiler?
  11. How are 2D / 3D CUDA blocks divided into warps?
  12. What is a bank conflict? (Doing Cuda/OpenCL programming)
  13. nvidia-smi Volatile GPU-Utilization explanation?
  14. How do I use Nvidia Multi-process Service (MPS) to run multiple non-MPI CUDA applications?
  15. Can anyone provide sample code demonstrating the use of 16 bit floating point in cuda?
  16. Any particular function to initialize GPU other than the first cudaMalloc call?
  17. CUDA and nvcc: using the preprocessor to choose between float or double
  18. Copy an object to device?
  19. Polymorphism and derived classes in CUDA / CUDA Thrust
  20. CUDA: How to use -arch and -code and SM vs COMPUTE
  21. How do I select which GPU to run a job on?
  22. How can I check the progress of matrix multiplication?
  23. device function pointers
  24. How can I make tensorflow run on a GPU with capability 2.x?
  25. cudaMemset() – does it set bytes or integers?
  26. How to implement device side CUDA virtual functions?
  27. CUDA determining threads per block, blocks per grid
  28. How does CUDA assign device IDs to GPUs?
  29. CUDA function pointers
  30. Using an array of device function pointers

Leave a Comment