Feat[model support]: Qwen Image — full pipeline with edit, generate LoRA, GGUF, quantization, and UI (#9000)

Author: lstein

invokeai/app/api/dependencies.py

@@ -51,6 +51,7 @@
     CogView4ConditioningInfo,
     ConditioningFieldData,
     FLUXConditioningInfo,
+    QwenImageConditioningInfo,
     SD3ConditioningInfo,
     SDXLConditioningInfo,
     ZImageConditioningInfo,
@@ -141,6 +142,7 @@ def initialize(
                     SD3ConditioningInfo,
                     CogView4ConditioningInfo,
                     ZImageConditioningInfo,
+                    QwenImageConditioningInfo,
                     AnimaConditioningInfo,
                 ],
                 ephemeral=True,

invokeai/app/invocations/fields.py

@@ -171,6 +171,8 @@ class FieldDescriptions:
     sd3_model = "SD3 model (MMDiTX) to load"
     cogview4_model = "CogView4 model (Transformer) to load"
     z_image_model = "Z-Image model (Transformer) to load"
+    qwen_image_model = "Qwen Image Edit model (Transformer) to load"
+    qwen_vl_encoder = "Qwen2.5-VL tokenizer, processor and text/vision encoder"
     sdxl_main_model = "SDXL Main model (UNet, VAE, CLIP1, CLIP2) to load"
     sdxl_refiner_model = "SDXL Refiner Main Modde (UNet, VAE, CLIP2) to load"
     onnx_main_model = "ONNX Main model (UNet, VAE, CLIP) to load"
@@ -340,6 +342,12 @@ class ZImageConditioningField(BaseModel):
     )
 
 
+class QwenImageConditioningField(BaseModel):
+    """A Qwen Image Edit conditioning tensor primitive value"""
+
+    conditioning_name: str = Field(description="The name of conditioning tensor")
+
+
 class AnimaConditioningField(BaseModel):
     """An Anima conditioning tensor primitive value.
 

invokeai/app/invocations/metadata.py

@@ -166,6 +166,10 @@ def invoke(self, context: InvocationContext) -> MetadataOutput:
     "z_image_img2img",
     "z_image_inpaint",
     "z_image_outpaint",
+    "qwen_image_txt2img",
+    "qwen_image_img2img",
+    "qwen_image_inpaint",
+    "qwen_image_outpaint",
     "anima_txt2img",
     "anima_img2img",
     "anima_inpaint",

invokeai/app/invocations/model.py

@@ -72,6 +72,13 @@ class GlmEncoderField(BaseModel):
     text_encoder: ModelIdentifierField = Field(description="Info to load text_encoder submodel")
 
 
+class QwenVLEncoderField(BaseModel):
+    """Field for Qwen2.5-VL encoder used by Qwen Image Edit models."""
+
+    tokenizer: ModelIdentifierField = Field(description="Info to load tokenizer submodel")
+    text_encoder: ModelIdentifierField = Field(description="Info to load text_encoder submodel")
+
+
 class Qwen3EncoderField(BaseModel):
     """Field for Qwen3 text encoder used by Z-Image models."""
 

invokeai/app/invocations/primitives.py

@@ -25,6 +25,7 @@
     InputField,
     LatentsField,
     OutputField,
+    QwenImageConditioningField,
     SD3ConditioningField,
     TensorField,
     UIComponent,
@@ -474,6 +475,17 @@ def build(cls, conditioning_name: str) -> "ZImageConditioningOutput":
         return cls(conditioning=ZImageConditioningField(conditioning_name=conditioning_name))
 
 
+@invocation_output("qwen_image_conditioning_output")
+class QwenImageConditioningOutput(BaseInvocationOutput):
+    """Base class for nodes that output a Qwen Image Edit conditioning tensor."""
+
+    conditioning: QwenImageConditioningField = OutputField(description=FieldDescriptions.cond)
+
+    @classmethod
+    def build(cls, conditioning_name: str) -> "QwenImageConditioningOutput":
+        return cls(conditioning=QwenImageConditioningField(conditioning_name=conditioning_name))
+
+
 @invocation_output("anima_conditioning_output")
 class AnimaConditioningOutput(BaseInvocationOutput):
     """Base class for nodes that output an Anima text conditioning tensor."""