feat: add LiteLLM integration (v1.8.0)

CHANGELOG.md

@@ -2,6 +2,18 @@
 
 All notable changes to `uipath_llm_client` (core package) will be documented in this file.
 
+## [1.8.0] - 2026-04-04
+
+### Added
+- `UiPathLiteLLM` — LiteLLM client wrapper for routing completions and embeddings through UiPath LLM Gateway
+  - Sync/async completions via `completion()` / `acompletion()`
+  - Sync/async embeddings via `embedding()` / `aembedding()`
+  - Injects UiPath-configured OpenAI client as HTTP transport
+- Added `litellm` optional dependency (`uipath-llm-client[litellm]`)
+
+### Changed
+- Bumped dependencies: `uipath-platform>=0.1.18`, `anthropic>=0.89.0`, `litellm>=1.83.1`
+
 ## [1.7.0] - 2026-04-03
 
 ### Added

packages/uipath_langchain_client/CHANGELOG.md

@@ -2,11 +2,17 @@
 
 All notable changes to `uipath_langchain_client` will be documented in this file.
 
-## [1.7.1] - 2026-04-04
+## [1.8.0] - 2026-04-04
 
 ### Added
+- `UiPathChatLiteLLM` — LangChain chat model wrapping LiteLLM routed through UiPath LLM Gateway
+- `UiPathLiteLLMEmbeddings` — LangChain embeddings wrapping LiteLLM routed through UiPath LLM Gateway
+- Added `litellm` optional dependency (`uipath-langchain-client[litellm]`)
 - `custom_class` parameter in `get_chat_model()` and `get_embedding_model()` factory functions to allow instantiating a user-provided class instead of the auto-detected one
 
+### Changed
+- Bumped dependencies: `langchain>=1.2.15`, `anthropic>=0.89.0`, `langchain-aws>=1.4.3`, `langchain-litellm>=0.6.4`
+
 ## [1.7.0] - 2026-04-03
 
 ### Added

packages/uipath_langchain_client/pyproject.toml

@@ -5,35 +5,38 @@ dynamic = ["version"]
 readme = "README.md"
 requires-python = ">=3.11"
 dependencies = [
-    "langchain>=1.2.13",
-    "uipath-llm-client>=1.7.0",
+    "langchain>=1.2.15",
+    "uipath-llm-client>=1.8.0",
 ]
 
 [project.optional-dependencies]
 openai = [
-    "langchain-openai>=1.1.11",
+    "langchain-openai>=1.1.12",
 ]
 google = [
     "langchain-google-genai>=4.2.1",
 ]
 anthropic = [
     "langchain-anthropic>=1.4.0",
-    "anthropic[bedrock,vertex]>=0.86.0",
+    "anthropic[bedrock,vertex]>=0.89.0",
 ]
 aws = [
-    "langchain-aws[anthropic]>=1.4.1",
+    "langchain-aws[anthropic]>=1.4.3",
 ]
 vertexai = [
     "langchain-google-vertexai>=3.2.2",
 ]
 azure = [
-    "langchain-azure-ai>=1.1.1",
+    "langchain-azure-ai>=1.2.1",
 ]
 fireworks = [
     "langchain-fireworks>=1.1.0",
 ]
+litellm = [
+    "langchain-litellm>=0.6.4",
+]
 all = [
-    "uipath-langchain-client[openai,aws,google,anthropic,azure,vertexai,fireworks]"
+    "uipath-langchain-client[openai,aws,google,anthropic,azure,vertexai,fireworks,litellm]"
 ]
 
 [build-system]

packages/uipath_langchain_client/src/uipath_langchain_client/__version__.py

@@ -1,3 +1,3 @@
 __title__ = "UiPath LangChain Client"
 __description__ = "A Python client for interacting with UiPath's LLM services via LangChain."
-__version__ = "1.7.1"
+__version__ = "1.8.0"

packages/uipath_langchain_client/src/uipath_langchain_client/clients/litellm/__init__.py

@@ -0,0 +1,7 @@
+from uipath_langchain_client.clients.litellm.chat_models import UiPathChatLiteLLM
+from uipath_langchain_client.clients.litellm.embeddings import UiPathLiteLLMEmbeddings
+
+__all__ = [
+    "UiPathChatLiteLLM",
+    "UiPathLiteLLMEmbeddings",
+]

packages/uipath_langchain_client/src/uipath_langchain_client/clients/litellm/chat_models.py

@@ -0,0 +1,82 @@
+from typing import Any, Optional, Self
+
+from pydantic import Field, model_validator
+
+from uipath_langchain_client.base_client import UiPathBaseChatModel
+
+try:
+    from langchain_litellm import ChatLiteLLM
+    from openai import AsyncOpenAI, OpenAI
+except ImportError as e:
+    raise ImportError(
+        "The 'litellm' extra is required to use UiPathChatLiteLLM. "
+        "Install it with: uv add uipath-langchain-client[litellm]"
+    ) from e
+
+from langchain_core.callbacks import (
+    AsyncCallbackManagerForLLMRun,
+    CallbackManagerForLLMRun,
+)
+
+
+class UiPathChatLiteLLM(UiPathBaseChatModel, ChatLiteLLM):  # type: ignore[override]
+    """LiteLLM chat model routed through UiPath LLM Gateway.
+
+    Combines UiPath's authentication and routing with LiteLLM's
+    multi-provider chat interface. The api_config must be provided
+    by the caller — provider routing is not auto-detected from model names.
+    """
+
+    # Override ChatLiteLLM's model field to align with UiPathBaseLLMClient.model_name (alias="model")
+    model: str = Field(default="", alias="model_name")
+
+    _uipath_openai_client: Optional[OpenAI] = None
+    _uipath_async_openai_client: Optional[AsyncOpenAI] = None
+
+    @model_validator(mode="after")
+    def setup_uipath_client(self) -> Self:
+        base_url = str(self.uipath_sync_client.base_url).rstrip("/")
+
+        self._uipath_openai_client = OpenAI(
+            api_key="PLACEHOLDER",
+            max_retries=0,
+            http_client=self.uipath_sync_client,
+            base_url=base_url,
+        )
+        self._uipath_async_openai_client = AsyncOpenAI(
+            api_key="PLACEHOLDER",
+            max_retries=0,
+            http_client=self.uipath_async_client,
+            base_url=base_url,
+        )
+        return self
+
+    def completion_with_retry(
+        self, run_manager: CallbackManagerForLLMRun | None = None, **kwargs: Any
+    ) -> Any:
+        """Override to inject UiPath OpenAI client into litellm calls."""
+        from langchain_litellm.chat_models.litellm import _create_retry_decorator
+
+        kwargs["client"] = self._uipath_openai_client
+        retry_decorator = _create_retry_decorator(self, run_manager=run_manager)
+
+        @retry_decorator
+        def _completion_with_retry(**kwargs: Any) -> Any:
+            return self.client.completion(**kwargs)
+
+        return _completion_with_retry(**kwargs)
+
+    async def acompletion_with_retry(
+        self, run_manager: AsyncCallbackManagerForLLMRun | None = None, **kwargs: Any
+    ) -> Any:
+        """Override to inject UiPath async OpenAI client into litellm calls."""
+        from langchain_litellm.chat_models.litellm import _create_retry_decorator
+
+        kwargs["client"] = self._uipath_async_openai_client
+        retry_decorator = _create_retry_decorator(self, run_manager=run_manager)
+
+        @retry_decorator
+        async def _completion_with_retry(**kwargs: Any) -> Any:
+            return await self.client.acompletion(**kwargs)
+
+        return await _completion_with_retry(**kwargs)

packages/uipath_langchain_client/src/uipath_langchain_client/clients/litellm/embeddings.py

@@ -0,0 +1,75 @@
+from typing import Any, Self
+
+from pydantic import Field, model_validator
+
+from uipath_langchain_client.base_client import UiPathBaseEmbeddings
+
+try:
+    from langchain_litellm import LiteLLMEmbeddings
+    from openai import AsyncOpenAI, OpenAI
+except ImportError as e:
+    raise ImportError(
+        "The 'litellm' extra is required to use UiPathLiteLLMEmbeddings. "
+        "Install it with: uv add uipath-langchain-client[litellm]"
+    ) from e
+
+
+class UiPathLiteLLMEmbeddings(UiPathBaseEmbeddings, LiteLLMEmbeddings):  # type: ignore[override]
+    """LiteLLM embeddings routed through UiPath LLM Gateway.
+
+    Combines UiPath's authentication and routing with LiteLLM's
+    multi-provider embeddings interface. The api_config must be provided
+    by the caller.
+    """
+
+    # Override LiteLLMEmbeddings' model field to align with UiPathBaseLLMClient.model_name
+    model: str = Field(default="", alias="model_name")
+
+    _uipath_openai_client: OpenAI | None = None
+    _uipath_async_openai_client: AsyncOpenAI | None = None
+
+    @model_validator(mode="after")
+    def setup_uipath_client(self) -> Self:
+        base_url = str(self.uipath_sync_client.base_url).rstrip("/")
+
+        self._uipath_openai_client = OpenAI(
+            api_key="PLACEHOLDER",
+            max_retries=0,
+            http_client=self.uipath_sync_client,
+            base_url=base_url,
+        )
+        self._uipath_async_openai_client = AsyncOpenAI(
+            api_key="PLACEHOLDER",
+            max_retries=0,
+            http_client=self.uipath_async_client,
+            base_url=base_url,
+        )
+        return self
+
+    def _embedding_with_retry(self, **kwargs: Any) -> Any:
+        """Override to inject UiPath OpenAI client into litellm embedding calls."""
+        import litellm
+        from langchain_litellm.embeddings.litellm import _create_retry_decorator
+
+        kwargs["client"] = self._uipath_openai_client
+        retry_decorator = _create_retry_decorator(self.max_retries)
+
+        @retry_decorator
+        def _embed() -> Any:
+            return litellm.embedding(**kwargs)
+
+        return _embed()
+
+    async def _aembedding_with_retry(self, **kwargs: Any) -> Any:
+        """Override to inject UiPath async OpenAI client into litellm embedding calls."""
+        import litellm
+        from langchain_litellm.embeddings.litellm import _create_retry_decorator
+
+        kwargs["client"] = self._uipath_async_openai_client
+        retry_decorator = _create_retry_decorator(self.max_retries)
+
+        @retry_decorator
+        async def _aembed() -> Any:
+            return await litellm.aembedding(**kwargs)
+
+        return await _aembed()

pyproject.toml

@@ -9,7 +9,7 @@ dependencies = [
     "tenacity>=9.1.4",
     "pydantic>=2.12.5",
     "pydantic-settings>=2.13.1",
-    "uipath-platform>=0.1.2",
+    "uipath-platform>=0.1.18",
 ]
 
 authors = [
@@ -19,16 +19,19 @@ authors = [
 
 [project.optional-dependencies]
 openai = [
-    "openai>=2.29.0",
+    "openai>=2.30.0",
 ]
 google = [
-    "google-genai>=1.68.0",
+    "google-genai>=1.70.0",
 ]
 anthropic = [
-    "anthropic>=0.86.0",
+    "anthropic>=0.89.0",
+]
+litellm = [
+    "litellm>=1.83.1",
 ]
 all = [
-    "uipath-llm-client[openai,google,anthropic]",
+    "uipath-llm-client[openai,google,anthropic,litellm]",
 ]
 
 [dependency-groups]
@@ -38,10 +41,10 @@ dev = [
     "pytest-asyncio>=1.3.0",
     "pytest-recording>=0.13.4",
     "pyright>=1.1.408",
-    "ruff>=0.15.7",
+    "ruff>=0.15.9",
+    "openinference-instrumentation-langchain>=0.1.62",
     "uipath-llm-client[all]",
     "uipath_langchain_client[all]",
-    "openinference-instrumentation-langchain>=0.1.61",
 ]
 
 [tool.uv.workspace]

src/uipath/llm_client/__version__.py

@@ -1,3 +1,3 @@
 __title__ = "UiPath LLM Client"
 __description__ = "A Python client for interacting with UiPath's LLM services."
-__version__ = "1.7.0"
+__version__ = "1.8.0"

src/uipath/llm_client/clients/litellm/__init__.py

@@ -0,0 +1,5 @@
+from uipath.llm_client.clients.litellm.client import UiPathLiteLLM
+
+__all__ = [
+    "UiPathLiteLLM",
+]