search

TheLastBen / fast-stable-diffusion

fast-stable-diffusion + DreamBooth
MIT License
7.53k stars 1.31k forks source link

Premium GPU batch img2img error #1244

Open Airdawg opened 1 year ago

Airdawg commented 1 year ago

Standard GPU is able to generate img2img batches but Premium GPU isn't working with GPU batch greater than 1.

TypeError: fwd(): incompatible function arguments. The following argument types are supported: 1. (arg0: at::Tensor, arg1: at::Tensor, arg2: at::Tensor, arg3: at::Tensor, arg4: at::Tensor, arg5: int, arg6: int, arg7: float, arg8: float, arg9: bool, arg10: bool, arg11: bool, arg12: Optional[at::Generator]) -> List[at::Tensor] Invoked with: tensor([[[ 8.7842e-01, -1.5371e+00, -2.8168e-02, ..., 8.3203e-01, 2.6875e+00, 4.3970e-01], [-7.5195e-01, -8.1787e-01, 1.5759e-01, ..., 1.5576e+00, 2.8281e+00, -6.4111e-01], [ 1.0801e+00, -1.0859e+00, 8.8965e-01, ..., 9.9463e-01, -1.5166e+00, 2.6343e-01], ..., [ 5.7910e-01, 1.0020e+00, 1.1084e+00, ..., -4.6997e-01, -9.2334e-01, -1.5881e-01], [ 3.2178e-01, -1.1597e-01, 8.2422e-01, ...,

TheLastBen commented 1 year ago

that must be an xformers issue, I'll recompile the lib later

Airdawg commented 1 year ago

that must be an xformers issue, I'll recompile the lib later

Tried again today. Standard GPU on colab is working but premium is now giving back this error:

FATAL: this function is for sm80, but was built for sm750

TheLastBen commented 1 year ago

using the latest colab ? https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast_stable_diffusion_AUTOMATIC1111.ipynb

https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast-DreamBooth.ipynb