Load float32 model

app.py

@@ -135,7 +135,7 @@ def inference(
         raise gr.Error("cuda is not available")
 
     device = torch.device("cuda")
-    model.to(device)
+    model.to(device).half()
 
     seed_everything(seed)
     start_code = torch.randn([len(prompts), 4, 128, 128], device=device)
@@ -162,7 +162,7 @@ def inference(
     register_attention_editor_diffusers(model, editor)
     images = model(prompts, latents=start_code, guidance_scale=classifier_free_guidance_scale).images
     unregister_attention_editor_diffusers(model)
-    model.to(torch.device("cpu"))
+    model.double().to(torch.device("cpu"))
     return images
 
 
@@ -256,7 +256,7 @@ def main():
     nltk.download("averaged_perceptron_tagger")
 
     scheduler = DDIMScheduler(beta_start=0.00085, beta_end=0.012, beta_schedule="scaled_linear", clip_sample=False, set_alpha_to_one=False)
-    model = StableDiffusionXLPipeline.from_pretrained(MODEL_PATH, scheduler=scheduler, torch_dtype=torch.float16)
+    model = StableDiffusionXLPipeline.from_pretrained(MODEL_PATH, scheduler=scheduler)
     model.unet.set_attn_processor(AttnProcessor2_0())
     model.enable_sequential_cpu_offload()