Update pipeline.py

pipeline.py

@@ -108,7 +108,7 @@ class FluxWithCFGPipeline(DiffusionPipeline, FluxLoraLoaderMixin, FromSingleFile
         self,
         prompt: Union[str, List[str]] = None,
         num_images_per_prompt: int = 1,
-        max_sequence_length: int = 512,
+        max_sequence_length: int = 256,
         device: Optional[torch.device] = None,
         dtype: Optional[torch.dtype] = None,
     ):
@@ -154,7 +154,6 @@ class FluxWithCFGPipeline(DiffusionPipeline, FluxLoraLoaderMixin, FromSingleFile
         self,
         prompt: Union[str, List[str]],
         num_images_per_prompt: int = 1,
-        max_sequence_length: int = 512,
         device: Optional[torch.device] = None,
     ):
         device = device or self._execution_device
@@ -180,7 +179,7 @@ class FluxWithCFGPipeline(DiffusionPipeline, FluxLoraLoaderMixin, FromSingleFile
                 "The following part of your input was truncated because CLIP can only handle sequences up to"
                 f" {self.tokenizer_max_length} tokens: {removed_text}"
             )
-        prompt_embeds = self.text_encoder(text_input_ids.to(device), output_hidden_states=False)
+        prompt_embeds = self.text_encoder(text_input_ids.to(device), output_hidden_states=True)
 
         # Use pooled output of CLIPTextModel
         prompt_embeds = prompt_embeds.pooler_output
@@ -190,7 +189,7 @@ class FluxWithCFGPipeline(DiffusionPipeline, FluxLoraLoaderMixin, FromSingleFile
 
         # duplicate text embeddings for each generation per prompt, using mps friendly method
         prompt_embeds = prompt_embeds.repeat(1, num_images_per_prompt)
-        prompt_embeds = prompt_embeds.view(batch_size * num_images_per_prompt, -1)
+        prompt_embeds = prompt_embeds.view(batch_size * num_images_per_prompt, seq_len, -1)
 
         return prompt_embeds
 
@@ -273,7 +272,6 @@ class FluxWithCFGPipeline(DiffusionPipeline, FluxLoraLoaderMixin, FromSingleFile
                 prompt=negative_prompt,
                 device=device,
                 num_images_per_prompt=num_images_per_prompt,
-                max_sequence_length=max_sequence_length,
             )
             
             t5_negative_prompt_embed = self._get_t5_prompt_embeds(