[Feat] Support Qianfan PP-StructureV3 MCP server (#17329)

* Support Qianfan PP-StructureV3 MCP server

* Bump version to 0.4.1

Author: Bobholamovic

mcp_server/paddleocr_mcp/__main__.py

@@ -123,6 +123,21 @@ def _validate_args(args: argparse.Namespace) -> None:
                 file=sys.stderr,
             )
             sys.exit(2)
+        elif args.ppocr_source == "qianfan":
+            if not args.qianfan_api_key:
+                print("Error: The Qianfan API key is required.", file=sys.stderr)
+                print(
+                    "Please either set `--qianfan_api_key` or set the environment variable "
+                    "`PADDLEOCR_MCP_QIANFAN_API_KEY`.",
+                    file=sys.stderr,
+                )
+                sys.exit(2)
+            if args.pipeline not in ("PaddleOCR-VL", "PP-StructureV3"):
+                print(
+                    f"{repr(args.pipeline)} is currently not supported when using the {repr(args.ppocr_source)} source.",
+                    file=sys.stderr,
+                )
+                sys.exit(2)
 
 
 async def async_main() -> None:

mcp_server/paddleocr_mcp/pipelines.py

@@ -655,7 +655,7 @@ async def _format_output(
 
 class _LayoutParsingHandler(SimpleInferencePipelineHandler):
     def _get_service_endpoint(self) -> str:
-        return "layout-parsing"
+        return "layout-parsing" if self._ppocr_source != "qianfan" else "paddleocr"
 
     def _transform_local_kwargs(self, kwargs: Dict[str, Any]) -> Dict[str, Any]:
         return {
@@ -869,6 +869,12 @@ def _create_local_engine(self) -> Any:
             device=self._device,
         )
 
+    def _transform_service_kwargs(self, kwargs: Dict[str, Any]) -> Dict[str, Any]:
+        kwargs = super()._transform_service_kwargs(kwargs)
+        if self._ppocr_source == "qianfan":
+            kwargs["model"] = "pp-structurev3"
+        return kwargs
+
 
 class PaddleOCRVLHandler(_LayoutParsingHandler):
     def register_tools(self, mcp: FastMCP) -> None:
@@ -908,6 +914,12 @@ def _create_local_engine(self) -> Any:
             device=self._device,
         )
 
+    def _transform_service_kwargs(self, kwargs: Dict[str, Any]) -> Dict[str, Any]:
+        kwargs = super()._transform_service_kwargs(kwargs)
+        if self._ppocr_source == "qianfan":
+            kwargs["model"] = "paddleocr-vl-0.9b"
+        return kwargs
+
 
 _PIPELINE_HANDLERS: Dict[str, Type[PipelineHandler]] = {
     "OCR": OCRHandler,

mcp_server/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "paddleocr_mcp"
-version = "0.3.0"
+version = "0.4.1"
 requires-python = ">=3.10"
 dependencies = [
     "mcp>=1.5.0",