dataautogpt3

/

OpenDalleV1.1

like 473

from diffusers import AutoPipelineForText2Image
import torch
        
pipeline = AutoPipelineForText2Image.from_pretrained('dataautogpt3/OpenDalleV1.1', torch_dtype=torch.float16).to('mps')        
image = pipeline('black fluffy gorgeous dangerous cat animal creature, large orange eyes, big fluffy ears, piercing gaze, full moon, dark ambiance, best quality, extremely detailed').images[0]

changed cuda to mps, and get the error below

Traceback (most recent call last):
  File "/private/tmp/opendallev1.1/launch_opendallev1.1.py", line 8, in <module>
    image = pipeline(t).images[0]
            ^^^^^^^^^^^
  File "/private/tmp/opendallev1.1/venv/lib/python3.11/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/private/tmp/opendallev1.1/venv/lib/python3.11/site-packages/diffusers/pipelines/stable_diffusion_xl/pipeline_stable_diffusion_xl.py", line 1113, in __call__
    timesteps, num_inference_steps = retrieve_timesteps(self.scheduler, num_inference_steps, device, timesteps)
                                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/private/tmp/opendallev1.1/venv/lib/python3.11/site-packages/diffusers/pipelines/stable_diffusion_xl/pipeline_stable_diffusion_xl.py", line 144, in retrieve_timesteps
    scheduler.set_timesteps(num_inference_steps, device=device, **kwargs)
  File "/private/tmp/opendallev1.1/venv/lib/python3.11/site-packages/diffusers/schedulers/scheduling_k_dpm_2_ancestral_discrete.py", line 280, in set_timesteps
    timesteps = torch.from_numpy(timesteps).to(device)
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
TypeError: Cannot convert a MPS Tensor to float64 dtype as the MPS framework doesn't support float64. Please use float32 instead.

Package            Version
------------------ ----------
accelerate         0.25.0
diffusers          0.25.0
torch              2.1.2
transformers       4.36.2