[Feature] Added Xlab Controlnet support (#11249)

update

src/diffusers/pipelines/flux/pipeline_flux_controlnet_image_to_image.py

@@ -800,6 +800,9 @@ class FluxControlNetImg2ImgPipeline(DiffusionPipeline, FluxLoraLoaderMixin, From
             )
             height, width = control_image.shape[-2:]
 
+            # xlab controlnet has a input_hint_block and instantx controlnet does not
+            controlnet_blocks_repeat = False if self.controlnet.input_hint_block is None else True
+            if self.controlnet.input_hint_block is None:
                 control_image = retrieve_latents(self.vae.encode(control_image), generator=generator)
                 control_image = (control_image - self.vae.config.shift_factor) * self.vae.config.scaling_factor
 
@@ -819,7 +822,9 @@ class FluxControlNetImg2ImgPipeline(DiffusionPipeline, FluxLoraLoaderMixin, From
         elif isinstance(self.controlnet, FluxMultiControlNetModel):
             control_images = []
 
-            for control_image_ in control_image:
+            # xlab controlnet has a input_hint_block and instantx controlnet does not
+            controlnet_blocks_repeat = False if self.controlnet.nets[0].input_hint_block is None else True
+            for i, control_image_ in enumerate(control_image):
                 control_image_ = self.prepare_image(
                     image=control_image_,
                     width=width,
@@ -831,6 +836,7 @@ class FluxControlNetImg2ImgPipeline(DiffusionPipeline, FluxLoraLoaderMixin, From
                 )
                 height, width = control_image_.shape[-2:]
 
+                if self.controlnet.nets[0].input_hint_block is None:
                     control_image_ = retrieve_latents(self.vae.encode(control_image_), generator=generator)
                     control_image_ = (control_image_ - self.vae.config.shift_factor) * self.vae.config.scaling_factor
 
@@ -955,6 +961,7 @@ class FluxControlNetImg2ImgPipeline(DiffusionPipeline, FluxLoraLoaderMixin, From
                     img_ids=latent_image_ids,
                     joint_attention_kwargs=self.joint_attention_kwargs,
                     return_dict=False,
+                    controlnet_blocks_repeat=controlnet_blocks_repeat,
                 )[0]
 
                 latents_dtype = latents.dtype