Merge branch 'main' into fix_requests.get

Author: kghamilton89

docs/source/en/api/pipelines/wan.md

@@ -307,6 +307,9 @@ def load_lora_adapter(self, pretrained_model_name_or_path_or_dict, prefix="trans
             try:
                 inject_adapter_in_model(lora_config, self, adapter_name=adapter_name, **peft_kwargs)
                 incompatible_keys = set_peft_model_state_dict(self, state_dict, adapter_name, **peft_kwargs)
+                # Set peft config loaded flag to True if module has been successfully injected and incompatible keys retrieved
+                if not self._hf_peft_config_loaded:
+                    self._hf_peft_config_loaded = True
             except Exception as e:
                 # In case `inject_adapter_in_model()` was unsuccessful even before injecting the `peft_config`.
                 if hasattr(self, "peft_config"):

src/diffusers/loaders/peft.py

@@ -282,6 +282,7 @@ def from_single_file(cls, pretrained_model_link_or_path_or_dict: Optional[str] =
         if quantization_config is not None:
             hf_quantizer = DiffusersAutoQuantizer.from_config(quantization_config)
             hf_quantizer.validate_environment()
+            torch_dtype = hf_quantizer.update_torch_dtype(torch_dtype)
 
         else:
             hf_quantizer = None

src/diffusers/loaders/single_file_model.py

@@ -273,7 +273,7 @@ def forward(
                 hidden_states = hidden_states.reshape(-1, hidden_states.shape[-2], hidden_states.shape[-1])
 
                 if i == 0 and num_frame > 1:
-                    hidden_states = hidden_states + self.temp_pos_embed
+                    hidden_states = hidden_states + self.temp_pos_embed.to(hidden_states.dtype)
 
                 if torch.is_grad_enabled() and self.gradient_checkpointing:
                     hidden_states = self._gradient_checkpointing_func(

src/diffusers/models/transformers/latte_transformer_3d.py

@@ -220,8 +220,13 @@ def _get_t5_prompt_embeds(
 
         return prompt_embeds
 
-    def encode_image(self, image: PipelineImageInput):
-        image = self.image_processor(images=image, return_tensors="pt").to(self.device)
+    def encode_image(
+        self,
+        image: PipelineImageInput,
+        device: Optional[torch.device] = None,
+    ):
+        device = device or self._execution_device
+        image = self.image_processor(images=image, return_tensors="pt").to(device)
         image_embeds = self.image_encoder(**image, output_hidden_states=True)
         return image_embeds.hidden_states[-2]
 
@@ -587,7 +592,7 @@ def __call__(
         if negative_prompt_embeds is not None:
             negative_prompt_embeds = negative_prompt_embeds.to(transformer_dtype)
 
-        image_embeds = self.encode_image(image)
+        image_embeds = self.encode_image(image, device)
         image_embeds = image_embeds.repeat(batch_size, 1, 1)
         image_embeds = image_embeds.to(transformer_dtype)
 

src/diffusers/pipelines/wan/pipeline_wan_i2v.py

@@ -90,13 +90,16 @@ class Base8bitTests(unittest.TestCase):
 
     def get_dummy_inputs(self):
         prompt_embeds = load_pt(
-            "https://huggingface.co/datasets/hf-internal-testing/bnb-diffusers-testing-artifacts/resolve/main/prompt_embeds.pt"
+            "https://huggingface.co/datasets/hf-internal-testing/bnb-diffusers-testing-artifacts/resolve/main/prompt_embeds.pt",
+            map_location="cpu",
         )
         pooled_prompt_embeds = load_pt(
-            "https://huggingface.co/datasets/hf-internal-testing/bnb-diffusers-testing-artifacts/resolve/main/pooled_prompt_embeds.pt"
+            "https://huggingface.co/datasets/hf-internal-testing/bnb-diffusers-testing-artifacts/resolve/main/pooled_prompt_embeds.pt",
+            map_location="cpu",
         )
         latent_model_input = load_pt(
-            "https://huggingface.co/datasets/hf-internal-testing/bnb-diffusers-testing-artifacts/resolve/main/latent_model_input.pt"
+            "https://huggingface.co/datasets/hf-internal-testing/bnb-diffusers-testing-artifacts/resolve/main/latent_model_input.pt",
+            map_location="cpu",
         )
 
         input_dict_for_transformer = {

tests/quantization/bnb/test_mixed_int8.py

@@ -57,7 +57,7 @@ def test_gguf_linear_layers(self):
             if isinstance(module, torch.nn.Linear) and hasattr(module.weight, "quant_type"):
                 assert module.weight.dtype == torch.uint8
                 if module.bias is not None:
-                    assert module.bias.dtype == torch.float32
+                    assert module.bias.dtype == self.torch_dtype
 
     def test_gguf_memory_usage(self):
         quantization_config = GGUFQuantizationConfig(compute_dtype=self.torch_dtype)