feat: Add watsonx embedding component (#7292)

* Add draft watsonx component

* feat: improved logic for embedding

* Add small changes to the ibm embedding component

* Fix icon and logger mode

---------

Co-authored-by: Thiago Araujo <thirauj@br.ibm.com>
Co-authored-by: Giovanni-Galatro <giovanni.galatro@ibm.com>
Co-authored-by: Jordan Frazier <122494242+jordanrfrazier@users.noreply.github.com>
Co-authored-by: galatro <galatrogiovanni@gmail.com>

src/backend/base/langflow/components/embeddings/__init__.py

@@ -14,6 +14,7 @@ from .openai import OpenAIEmbeddingsComponent
 from .similarity import EmbeddingSimilarityComponent
 from .text_embedder import TextEmbedderComponent
 from .vertexai import VertexAIEmbeddingsComponent
+from .watsonx import WatsonxEmbeddingsComponent
 
 __all__ = [
     "AIMLEmbeddingsComponent",
@@ -32,4 +33,5 @@ __all__ = [
     "OpenAIEmbeddingsComponent",
     "TextEmbedderComponent",
     "VertexAIEmbeddingsComponent",
+    "WatsonxEmbeddingsComponent",
 ]

src/backend/base/langflow/components/embeddings/watsonx.py

@@ -0,0 +1,136 @@
+import logging
+from typing import Any
+
+import requests
+from ibm_watsonx_ai import APIClient, Credentials
+from ibm_watsonx_ai.metanames import EmbedTextParamsMetaNames
+from langchain_ibm import WatsonxEmbeddings
+from pydantic.v1 import SecretStr
+
+from langflow.base.embeddings.model import LCEmbeddingsModel
+from langflow.field_typing import Embeddings
+from langflow.io import BoolInput, DropdownInput, IntInput, SecretStrInput, StrInput
+from langflow.schema.dotdict import dotdict
+
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+
+class WatsonxEmbeddingsComponent(LCEmbeddingsModel):
+    display_name = "IBM watsonx.ai Embeddings"
+    description = "Generate embeddings using IBM watsonx.ai models."
+    icon = "WatsonxAI"
+    name = "WatsonxEmbeddingsComponent"
+
+    # models present in all the regions
+    _default_models = [
+        "sentence-transformers/all-minilm-l12-v2",
+        "ibm/slate-125m-english-rtrvr-v2",
+        "ibm/slate-30m-english-rtrvr-v2",
+        "intfloat/multilingual-e5-large",
+    ]
+
+    inputs = [
+        DropdownInput(
+            name="url",
+            display_name="watsonx API Endpoint",
+            info="The base URL of the API.",
+            value=None,
+            options=[
+                "https://us-south.ml.cloud.ibm.com",
+                "https://eu-de.ml.cloud.ibm.com",
+                "https://eu-gb.ml.cloud.ibm.com",
+                "https://au-syd.ml.cloud.ibm.com",
+                "https://jp-tok.ml.cloud.ibm.com",
+                "https://ca-tor.ml.cloud.ibm.com",
+            ],
+            real_time_refresh=True,
+        ),
+        StrInput(
+            name="project_id",
+            display_name="watsonx project id",
+        ),
+        SecretStrInput(
+            name="api_key",
+            display_name="API Key",
+            info="The API Key to use for the model.",
+            required=True,
+        ),
+        DropdownInput(
+            name="model_name",
+            display_name="Model Name",
+            options=[],
+            value=None,
+            dynamic=True,
+            required=True,
+        ),
+        IntInput(
+            name="truncate_input_tokens",
+            display_name="Truncate Input Tokens",
+            advanced=True,
+            value=200,
+        ),
+        BoolInput(
+            name="input_text",
+            display_name="Include the original text in the output",
+            value=True,
+            advanced=True,
+        ),
+    ]
+
+    @staticmethod
+    def fetch_models(base_url: str) -> list[str]:
+        """Fetch available models from the watsonx.ai API."""
+        try:
+            endpoint = f"{base_url}/ml/v1/foundation_model_specs"
+            params = {
+                "version": "2024-09-16",
+                "filters": "function_embedding,!lifecycle_withdrawn:and",
+            }
+            response = requests.get(endpoint, params=params, timeout=10)
+            response.raise_for_status()
+            data = response.json()
+            models = [model["model_id"] for model in data.get("resources", [])]
+            return sorted(models)
+        except Exception:
+            logger.exception("Error fetching models")
+            return WatsonxEmbeddingsComponent._default_models
+
+    def update_build_config(self, build_config: dotdict, field_value: Any, field_name: str | None = None):
+        """Update model options when URL or API key changes."""
+        logger.debug(
+            "Updating build config. Field name: %s, Field value: %s",
+            field_name,
+            field_value,
+        )
+
+        if field_name == "url" and field_value:
+            try:
+                models = self.fetch_models(base_url=build_config.url.value)
+                build_config.model_name.options = models
+                if build_config.model_name.value:
+                    build_config.model_name.value = models[0]
+                info_message = f"Updated model options: {len(models)} models found in {build_config.url.value}"
+                logger.info(info_message)
+            except Exception:
+                logger.exception("Error updating model options.")
+
+    def build_embeddings(self) -> Embeddings:
+        credentials = Credentials(
+            api_key=SecretStr(self.api_key).get_secret_value(),
+            url=self.url,
+        )
+
+        api_client = APIClient(credentials)
+
+        params = {
+            EmbedTextParamsMetaNames.TRUNCATE_INPUT_TOKENS: self.truncate_input_tokens,
+            EmbedTextParamsMetaNames.RETURN_OPTIONS: {"input_text": self.input_text},
+        }
+
+        return WatsonxEmbeddings(
+            model_id=self.model_name,
+            params=params,
+            watsonx_client=api_client,
+            project_id=self.project_id,
+        )