feat(apionly): add Alibaba Cloud Models

docs/features/external-models/alibabacloud.md

@@ -0,0 +1,49 @@
+---
+title: Alibaba Cloud DashScope
+---
+
+# :material-cloud-outline: Alibaba Cloud DashScope
+
+Invoke supports Alibaba Cloud's **DashScope** image generation service, giving access to the **Qwen Image** family and **Wan 2.6** text-to-image. Qwen Image is particularly strong at bilingual (Chinese / English) text rendering.
+
+## Getting an API Key
+
+1. Sign in to [Alibaba Cloud Model Studio](https://www.alibabacloud.com/en/product/modelstudio) (the international DashScope portal).
+2. Enable **DashScope** and activate the image generation models you plan to use.
+3. Create an API key from the **API Keys** section of the console.
+
+## Configuration
+
+Add your key to `api_keys.yaml` in your Invoke root directory:
+
+```yaml
+external_alibabacloud_api_key: "your-dashscope-api-key"
+
+# Optional — default is the international endpoint. Use the China endpoint if your account lives there:
+#   https://dashscope.aliyuncs.com
+external_alibabacloud_base_url: "https://dashscope-intl.aliyuncs.com"
+```
+
+Restart Invoke for the change to take effect.
+
+!!! info "International vs. China endpoints"
+    DashScope has separate international (`dashscope-intl.aliyuncs.com`) and China (`dashscope.aliyuncs.com`) deployments. Your API key only works on the deployment it was issued on — if you get authentication errors, check that `external_alibabacloud_base_url` matches.
+
+## Available Models
+
+| Model | Modes | Aspect Ratios | Batch | Notes |
+| --- | --- | --- | --- | --- |
+| **Qwen Image 2.0 Pro** | txt2img | 1:1, 4:3, 3:4, 16:9, 9:16 | up to 4 | Best quality, 2K output, excellent bilingual text. |
+| **Qwen Image 2.0** | txt2img | 1:1, 4:3, 3:4, 16:9, 9:16 | up to 4 | Faster / cheaper 2K sibling of 2.0 Pro. |
+| **Qwen Image Max** | txt2img | 1:1, 4:3, 3:4, 16:9, 9:16 | up to 4 | High quality at ~1.3K native size. |
+| **Qwen Image Edit Max** | img2img | 1:1, 4:3, 3:4, 16:9, 9:16 | up to 4 | Image editing with industrial / geometric reasoning. |
+| **Wan 2.6 Text-to-Image** | txt2img | 1:1, 4:3, 3:4, 16:9, 9:16 | up to 4 | Photorealistic T2I at 1K. |
+
+All models support **negative prompt** and **seed**.
+
+## Tips
+
+- **Bilingual prompts.** Qwen Image is unusually good at rendering Chinese text and mixed-language prompts — it's a strong choice when your prompt or desired output contains non-Latin script.
+- **Editing** is only supported by Qwen Image Edit Max. Use it for image-to-image workflows; the other Qwen Image models are text-to-image only.
+- **Batching** is capped at 4 images per request. Larger batches are split across multiple API calls.
+- **Costs** vary per model — Qwen Image 2.0 Pro is the most expensive, Qwen Image 2.0 the cheapest of the 2.0 family. Check Alibaba Cloud's pricing page before running large batches.

docs/src/generated/settings.json

@@ -724,6 +724,28 @@
       "type": "<class 'bool'>",
       "validation": {}
     },
+    {
+      "category": "EXTERNAL PROVIDERS",
+      "default": null,
+      "description": "API key for Alibaba Cloud DashScope image generation.",
+      "env_var": "INVOKEAI_EXTERNAL_ALIBABACLOUD_API_KEY",
+      "literal_values": [],
+      "name": "external_alibabacloud_api_key",
+      "required": false,
+      "type": "typing.Optional[str]",
+      "validation": {}
+    },
+    {
+      "category": "EXTERNAL PROVIDERS",
+      "default": null,
+      "description": "Base URL override for Alibaba Cloud DashScope image generation.",
+      "env_var": "INVOKEAI_EXTERNAL_ALIBABACLOUD_BASE_URL",
+      "literal_values": [],
+      "name": "external_alibabacloud_base_url",
+      "required": false,
+      "type": "typing.Optional[str]",
+      "validation": {}
+    },
     {
       "category": "EXTERNAL PROVIDERS",
       "default": null,

invokeai/app/api/dependencies.py

@@ -17,7 +17,7 @@
 from invokeai.app.services.download.download_default import DownloadQueueService
 from invokeai.app.services.events.events_fastapievents import FastAPIEventService
 from invokeai.app.services.external_generation.external_generation_default import ExternalGenerationService
-from invokeai.app.services.external_generation.providers import GeminiProvider, OpenAIProvider
+from invokeai.app.services.external_generation.providers import AlibabaCloudProvider, GeminiProvider, OpenAIProvider
 from invokeai.app.services.external_generation.startup import sync_configured_external_starter_models
 from invokeai.app.services.image_files.image_files_disk import DiskImageFileStorage
 from invokeai.app.services.image_records.image_records_sqlite import SqliteImageRecordStorage
@@ -161,6 +161,7 @@ def initialize(
         )
         external_generation = ExternalGenerationService(
             providers={
+                AlibabaCloudProvider.provider_id: AlibabaCloudProvider(app_config=configuration, logger=logger),
                 GeminiProvider.provider_id: GeminiProvider(app_config=configuration, logger=logger),
                 OpenAIProvider.provider_id: OpenAIProvider(app_config=configuration, logger=logger),
             },

invokeai/app/api/routers/app_info.py

@@ -97,6 +97,7 @@ class ExternalProviderConfigModel(BaseModel):
 
 
 EXTERNAL_PROVIDER_FIELDS: dict[str, tuple[str, str]] = {
+    "alibabacloud": ("external_alibabacloud_api_key", "external_alibabacloud_base_url"),
     "gemini": ("external_gemini_api_key", "external_gemini_base_url"),
     "openai": ("external_openai_api_key", "external_openai_base_url"),
 }

invokeai/app/invocations/external_image_generation.py

@@ -267,15 +267,43 @@ class GeminiImageGenerationInvocation(BaseExternalImageGenerationInvocation):
     mask_image: ImageField | None = InputField(default=None, description="Mask image for inpaint", ui_hidden=True)
 
     temperature: float | None = InputField(default=None, ge=0.0, le=2.0, description="Sampling temperature")
+    thinking_level: Literal["minimal", "high"] | None = InputField(
+        default=None, description="Thinking level for image generation"
+    )
 
     def _build_provider_options(self) -> dict[str, Any] | None:
         options: dict[str, Any] = {}
         if self.temperature is not None:
             options["temperature"] = self.temperature
+        if self.thinking_level is not None:
+            options["thinking_level"] = self.thinking_level
         return options or None
 
     def _build_output_provider_metadata(self) -> dict[str, Any]:
         metadata: dict[str, Any] = {}
         if self.temperature is not None:
             metadata["gemini_temperature"] = self.temperature
+        if self.thinking_level is not None:
+            metadata["gemini_thinking_level"] = self.thinking_level
         return metadata
+
+
+@invocation(
+    "alibabacloud_image_generation",
+    title="Alibaba Cloud DashScope Image Generation",
+    tags=["external", "generation", "alibabacloud", "dashscope"],
+    category="image",
+    version="1.0.0",
+)
+class AlibabaCloudImageGenerationInvocation(BaseExternalImageGenerationInvocation):
+    """Generate images using an Alibaba Cloud DashScope external model."""
+
+    provider_id = "alibabacloud"
+
+    model: ModelIdentifierField = InputField(
+        description=FieldDescriptions.main_model,
+        ui_model_base=[BaseModelType.External],
+        ui_model_type=[ModelType.ExternalImageGenerator],
+        ui_model_format=[ModelFormat.ExternalApi],
+        ui_model_provider_id=["alibabacloud"],
+    )

invokeai/app/services/config/config_default.py

@@ -32,9 +32,11 @@
 LOG_LEVEL = Literal["debug", "info", "warning", "error", "critical"]
 CONFIG_SCHEMA_VERSION = "4.0.2"
 EXTERNAL_PROVIDER_CONFIG_FIELDS = (
+    "external_alibabacloud_api_key",
+    "external_alibabacloud_base_url",
     "external_gemini_api_key",
-    "external_openai_api_key",
     "external_gemini_base_url",
+    "external_openai_api_key",
     "external_openai_base_url",
 )
 
@@ -120,6 +122,8 @@ class InvokeAIAppConfig(BaseSettings):
         allow_unknown_models: Allow installation of models that we are unable to identify. If enabled, models will be marked as `unknown` in the database, and will not have any metadata associated with them. If disabled, unknown models will be rejected during installation.
         multiuser: Enable multiuser support. When disabled, the application runs in single-user mode using a default system account with administrator privileges. When enabled, requires user authentication and authorization.
         strict_password_checking: Enforce strict password requirements. When True, passwords must contain uppercase, lowercase, and numbers. When False (default), any password is accepted but its strength (weak/moderate/strong) is reported to the user.
+        external_alibabacloud_api_key: API key for Alibaba Cloud DashScope image generation.
+        external_alibabacloud_base_url: Base URL override for Alibaba Cloud DashScope image generation.
         external_gemini_api_key: API key for Gemini image generation.
         external_openai_api_key: API key for OpenAI image generation.
         external_gemini_base_url: Base URL override for Gemini image generation.
@@ -223,6 +227,10 @@ class InvokeAIAppConfig(BaseSettings):
     strict_password_checking:      bool = Field(default=False,              description="Enforce strict password requirements. When True, passwords must contain uppercase, lowercase, and numbers. When False (default), any password is accepted but its strength (weak/moderate/strong) is reported to the user.")
 
     # EXTERNAL PROVIDERS
+    external_alibabacloud_api_key: Optional[str] = Field(default=None, description="API key for Alibaba Cloud DashScope image generation.")
+    external_alibabacloud_base_url: Optional[str] = Field(
+        default=None, description="Base URL override for Alibaba Cloud DashScope image generation."
+    )
     external_gemini_api_key: Optional[str] = Field(default=None, description="API key for Gemini image generation.")
     external_openai_api_key: Optional[str] = Field(default=None, description="API key for OpenAI image generation.")
     external_gemini_base_url: Optional[str] = Field(

invokeai/app/services/external_generation/providers/__init__.py

@@ -1,4 +1,5 @@
+from invokeai.app.services.external_generation.providers.alibabacloud import AlibabaCloudProvider
 from invokeai.app.services.external_generation.providers.gemini import GeminiProvider
 from invokeai.app.services.external_generation.providers.openai import OpenAIProvider
 
-__all__ = ["GeminiProvider", "OpenAIProvider"]
+__all__ = ["AlibabaCloudProvider", "GeminiProvider", "OpenAIProvider"]