UiPath LLM Client [v1.9.5]

[1.9.5] - 2026-04-21

Added

• utils.headers.UIPATH_DEFAULT_REQUEST_HEADERS public constant — the single source of truth for built-in gateway request headers (X-UiPath-LLMGateway-TimeoutSeconds=295, X-UiPath-LLMGateway-AllowFull4xxResponse=false). UiPathHttpxClient._default_headers now references this constant; the langchain base client reuses the same constant for its class_default_headers.

UiPath LLM Client [v1.9.4]

[1.9.4] - 2026-04-21

Changed

• Bumped dependency floors to the latest released versions: pydantic-settings>=2.14.0, uipath-platform>=0.1.32, google-genai>=1.73.1, anthropic>=0.96.0, litellm>=1.83.7.

UiPath LangChain Client [langchain-v1.9.5]

[1.9.5] - 2026-04-21

Changed

• UiPathBaseLLMClient.default_headers is now additive: caller-supplied headers are merged on top of a class-level class_default_headers (timeout and AllowFull4xxResponse policy) instead of replacing them. User values still win on key collisions. Previously, passing any default_headers={...} caused the built-in defaults to be dropped from self.default_headers (though the core httpx client's class defaults kept them on the wire).
• UiPathBaseLLMClient.class_default_headers now points at the shared uipath.llm_client.utils.headers.UIPATH_DEFAULT_REQUEST_HEADERS constant (single source of truth with core's UiPathHttpxClient._default_headers).
• Minimum uipath-llm-client bumped to 1.9.5 for the shared UIPATH_DEFAULT_REQUEST_HEADERS constant.

UiPath LangChain Client [langchain-v1.9.4]

[1.9.4] - 2026-04-21

Changed

• Bumped dependency floors to the latest released versions: langchain-openai>=1.1.15, langchain-google-genai>=4.2.2, langchain-anthropic>=1.4.1, anthropic[bedrock,vertex]>=0.96.0, langchain-aws[anthropic]>=1.4.4, langchain-azure-ai>=1.2.2.
• Minimum uipath-llm-client bumped to 1.9.4 to match the core dependency-floor release.

UiPath LangChain Client [langchain-v1.9.3]

[1.9.3] - 2026-04-20

Changed

• get_chat_model() now defaults to the OpenAI Responses API (ApiFlavor.RESPONSES) when discovery does not specify a flavor for an OpenAI chat model. Explicit api_flavor= on the call and BYOM-discovered flavors still take precedence. The LiteLLM client still defaults to chat-completions for OpenAI because LiteLLM 1.83.x drops the injected httpx client when its acompletion→aresponses bridge fires, which breaks async auth against the UiPath gateway.

UiPath LLM Client [v1.9.2]

[1.9.2] - 2026-04-17

Changed

• PlatformBaseSettings.build_auth_headers() now uses the header-name constants from uipath.platform.common.constants (lowercase canonical form). HTTP header names are case-insensitive so wire-level behavior is unchanged.
• UIPATH_PROCESS_KEY is now URL-encoded (urllib.parse.quote(..., safe="")) before being placed in X-UiPath-ProcessKey, matching the platform-wide convention.

Added

• HEADER_LICENSING_CONTEXT header populated dynamically from UiPathConfig.licensing_context when set.

UiPath LLM Client [v1.9.1]

[1.9.1] - 2026-04-17

Added

• utils.model_family.is_anthropic_model_name() helper and ANTHROPIC_MODEL_NAME_KEYWORDS tuple — name-based Claude detection for BYOM deployments where discovery does not expose modelFamily

Fixed

• UiPathLiteLLM now detects Claude-family models by name when modelFamily is unavailable (BYOM), correctly routing Bedrock/Vertex provider selection and default flavors

UiPath LLM Client [v1.9.0]

[1.9.0] - 2026-04-17

Added

• ModelFamily StrEnum constants (OPENAI, GOOGLE_GEMINI, ANTHROPIC_CLAUDE) for model family matching
• get_model_info() on UiPathBaseSettings — centralized model lookup with filtering by name, vendor, and BYO connection ID
• Discovery cache on get_available_models() keyed by settings properties, with refresh parameter to bypass

Changed

• get_available_models() is now a concrete cached method on the base class; subclasses implement _fetch_available_models() instead
• validate_byo_model() is now a default no-op on the base class (only LLMGateway overrides it) and is called automatically inside get_model_info()
• LiteLLM client uses get_model_info() instead of duplicating model discovery logic

UiPath LangChain Client [langchain-v1.9.2]

[1.9.2] - 2026-04-17

Changed

• Breaking: captured gateway headers are now exposed on AIMessage.response_metadata under the headers key (previously uipath_llmgateway_headers). Update any consumers that read this key.
• Minimum uipath-llm-client bumped to 1.9.2 for the platform-headers refactor and licensing-context support.

UiPath LangChain Client [langchain-v1.9.1]

[1.9.1] - 2026-04-17

Fixed

• Detect Anthropic-family models by additional name keywords (anthropic, opus, sonnet, haiku, mythos) alongside claude — applies to Bedrock INVOKE factory routing and the normalized client's empty tool-call content workaround. Uses the shared is_anthropic_model_name() helper from core 1.9.1.

Changed

• Minimum uipath-llm-client bumped to 1.9.1 for the shared is_anthropic_model_name() helper