Upgrading langchain 0.0 to 0.1
Breaking0.0.354 → 0.1.20Security2Platform3New5Changed20Fixed8Other1
SecurityLangChain through 0.1.10 allows ../ directory traversal by an actor who is able to control the final part of the path parameter in a load_chain call. This bypasses the intended behavior of loading configurations only from the hwchase17/langchain-hub GitHub repository. The outcome can be disclosure of an API key for a large language model online service, or remote code execution.GHSA-h59x-p739-982c · CVE-2024-28088 ↗
BreakingUpgrade verdict
3 platform requirement changes and 2 security fixes.
In plain terms
- release(nomic): 0.1.5 revert: "chore: remove ruff target-version" chore: remove ruff target-version chore: adress pytest-asyncio deprecation warnings + other nits feat(docs): improve devx, fix…
- Release openai version fix: bump deps for release release(deepseek): 0.1.4 fix(deepseek): convert tool output arrays to strings fix: bump lockfiles ruff: restore stacklevels, disable autofixing…
- release(prompty): 0.1.2 revert: "chore: remove ruff target-version" chore: remove ruff target-version chore: adress pytest-asyncio deprecation warnings + other nits chore: formatting across codebase…
- LangChain through 0.1.10 allows ../ directory traversal by an actor who is able to control the final part of the path parameter in a load_chain call. This bypasses the intended behavior of loading configurations only from the hwchase17/langchain-hub GitHub repository. The outcome can be disclosure of an API key for a large language model online service, or remote code execution.
Must act 5 items need code or config changes
Python requirement changed in 0.1.13: >=3.8.1,<4.0 → <4.0,>=3.8.1.
Adds langchain-text-splitters >=0.0.1,<0.1.
Drops jsonpatch.
Installs 39 packages in all (was 33).
What you must change
Security2
LangChain through 0.1.10 allows ../ directory traversal by an actor who is able to control the final part of the path parameter in a load_chain call. This bypasses the intended behavior of loading configurations only from the hwchase17/langchain-hub GitHub repository. The outcome can be disclosure of an API key for a large language model online service, or remote code execution.
langchain Server-Side Request Forgery vulnerability
Low · CVSS 3.7Platform3
release(nomic): 0.1.5 revert: "chore: remove ruff target-version" chore: remove ruff target-version chore: adress pytest-asyncio deprecation warnings + other nits feat(docs): improve devx, fix…
Makefileextra_bodypyupgradenumpynumpy>=2.1.0Release openai version fix: bump deps for release release(deepseek): 0.1.4 fix(deepseek): convert tool output arrays to strings fix: bump lockfiles ruff: restore stacklevels, disable autofixing…
reasoning_contentpyupgraderelease(prompty): 0.1.2 revert: "chore: remove ruff target-version" chore: remove ruff target-version chore: adress pytest-asyncio deprecation warnings + other nits chore: formatting across codebase…
Makefileextra_bodypyupgradeAPI changes
Generated by DepDelta from published Python source. Not mixed with prose notes.
- added:
export create_tool_calling_agent
- added:
field RunnableAgent.input_keys_arg: List[str] - added:
field RunnableAgent.return_keys_arg: List[str] - added:
field RunnableAgent.stream_runnable: bool - added:
field RunnableMultiActionAgent.input_keys_arg: List[str] - added:
field RunnableMultiActionAgent.return_keys_arg: List[str] - added:
field RunnableMultiActionAgent.stream_runnable: bool
- removed:
export SCOPES
- removed:
export MAX_RESPONSE_LENGTH
- added:
field ChatOutputParser.format_instructions: str
- added:
field ConvoOutputParser.format_instructions: str
- added:
field ConvoOutputParser.format_instructions: str
- added:
export format_to_tool_messages
- removed:
function format_to_openai_tool_messages(intermediate_steps: Sequence[Tuple[AgentAction, str]]) -> List[BaseMessage] - added:
export format_to_openai_tool_messages
- added:
function format_to_tool_messages(intermediate_steps: Sequence[Tuple[AgentAction, str]]) -> List[BaseMessage]
- changed:
function create_json_chat_agent(llm: BaseLanguageModel, tools: Sequence[BaseTool], prompt: ChatPromptTemplate) -> Runnable → function create_json_chat_agent(llm: BaseLanguageModel, tools: Sequence[BaseTool], prompt: ChatPromptTemplate, stop_sequence: Union[bool, List[str]]=…, tools_renderer: ToolsRenderer=…, template_tool_response: str=…) -> Runnable
- removed:
function get_all_tool_names() -> List[str] - removed:
function load_huggingface_tool(task_or_repo_id: str, model_repo_id: Optional[str]=…, token: Optional[str]=…, remote: bool=…, **kwargs: Any) -> BaseTool - removed:
function load_tools(tool_names: List[str], llm: Optional[BaseLanguageModel]=…, callbacks: Callbacks=…, **kwargs: Any) -> List[BaseTool]
- added:
field MRKLOutputParser.format_instructions: str
- changed:
field OpenAIAssistantRunnable.client: openai.OpenAI → field OpenAIAssistantRunnable.client: Any - changed:
static method OpenAIAssistantRunnable.create_assistant(name: str, instructions: str, tools: Sequence[Union[BaseTool, dict]], model: str, *, client: Optional[openai.OpenAI]=…, **kwargs: Any) -> OpenAIAssistantRunnable → static method OpenAIAssistantRunnable.create_assistant(name: str, instructions: str, tools: Sequence[Union[BaseTool, dict]], model: str, *, client: Optional[Union[openai.OpenAI, openai.AzureOpenAI]]=…, **kwargs: Any) -> OpenAIAssistantRunnable - added:
async method OpenAIAssistantRunnable.ainvoke(input: dict, config: Optional[RunnableConfig]=…, **kwargs: Any) -> OutputType - added:
async static method OpenAIAssistantRunnable.acreate_assistant(name: str, instructions: str, tools: Sequence[Union[BaseTool, dict]], model: str, *, async_client: Optional[Union[openai.AsyncOpenAI, openai.AsyncAzureOpenAI]]=…, **kwargs: Any) -> OpenAIAssistantRunnable - added:
field OpenAIAssistantRunnable.async_client: Any - added:
method OpenAIAssistantRunnable.validate_async_client(values: dict) -> dict
- added:
field AgentTokenBufferMemory.format_as_tools: bool
- changed:
static method OpenAIFunctionsAgent.create_prompt(system_message: Optional[SystemMessage]=…, extra_prompt_messages: Optional[List[BaseMessagePromptTemplate]]=…) -> BasePromptTemplate → static method OpenAIFunctionsAgent.create_prompt(system_message: Optional[SystemMessage]=…, extra_prompt_messages: Optional[List[BaseMessagePromptTemplate]]=…) -> ChatPromptTemplate - added:
field OpenAIFunctionsAgent.output_parser: Type[OpenAIFunctionsAgentOutputParser]
- added:
export ToolsAgentOutputParser
- removed:
class OpenAIToolAgentAction(AgentActionMessageLog) - removed:
field OpenAIToolAgentAction.tool_call_id: str - added:
variable OpenAIToolAgentAction
- added:
class ToolAgentAction(AgentActionMessageLog) - added:
class ToolsAgentOutputParser(MultiActionAgentOutputParser) - added:
field ToolAgentAction.tool_call_id: str - added:
function parse_ai_message_to_tool_action(message: BaseMessage) -> Union[List[AgentAction], AgentFinish] - added:
method ToolsAgentOutputParser.parse(text: str) -> Union[List[AgentAction], AgentFinish] - added:
method ToolsAgentOutputParser.parse_result(result: List[Generation], *, partial: bool=…) -> Union[List[AgentAction], AgentFinish]
- changed:
function create_react_agent(llm: BaseLanguageModel, tools: Sequence[BaseTool], prompt: BasePromptTemplate) -> Runnable → function create_react_agent(llm: BaseLanguageModel, tools: Sequence[BaseTool], prompt: BasePromptTemplate, output_parser: Optional[AgentOutputParser]=…, tools_renderer: ToolsRenderer=…, *, stop_sequence: Union[bool, List[str]]=…) -> Runnable
- changed:
function create_structured_chat_agent(llm: BaseLanguageModel, tools: Sequence[BaseTool], prompt: ChatPromptTemplate) -> Runnable → function create_structured_chat_agent(llm: BaseLanguageModel, tools: Sequence[BaseTool], prompt: ChatPromptTemplate, tools_renderer: ToolsRenderer=…, *, stop_sequence: Union[bool, List[str]]=…) -> Runnable
- added:
field StructuredChatOutputParser.format_instructions: str
- added:
function create_tool_calling_agent(llm: BaseLanguageModel, tools: Sequence[BaseTool], prompt: ChatPromptTemplate) -> Runnable
- changed:
function create_xml_agent(llm: BaseLanguageModel, tools: Sequence[BaseTool], prompt: BasePromptTemplate) -> Runnable → function create_xml_agent(llm: BaseLanguageModel, tools: Sequence[BaseTool], prompt: BasePromptTemplate, tools_renderer: ToolsRenderer=…, *, stop_sequence: Union[bool, List[str]]=…) -> Runnable
- added:
export AzureCosmosDBSemanticCache
- added:
export tracing_enabled
- removed:
class FileCallbackHandler(BaseCallbackHandler) - removed:
constructor FileCallbackHandler(filename: str, mode: str=…, color: Optional[str]=…) -> None - removed:
method FileCallbackHandler.on_agent_action(action: AgentAction, color: Optional[str]=…, **kwargs: Any) -> Any - removed:
method FileCallbackHandler.on_agent_finish(finish: AgentFinish, color: Optional[str]=…, **kwargs: Any) -> None - removed:
method FileCallbackHandler.on_chain_end(outputs: Dict[str, Any], **kwargs: Any) -> None - removed:
method FileCallbackHandler.on_chain_start(serialized: Dict[str, Any], inputs: Dict[str, Any], **kwargs: Any) -> None - removed:
method FileCallbackHandler.on_text(text: str, color: Optional[str]=…, end: str=…, **kwargs: Any) -> None - removed:
method FileCallbackHandler.on_tool_end(output: str, color: Optional[str]=…, observation_prefix: Optional[str]=…, llm_prefix: Optional[str]=…, **kwargs: Any) -> None - added:
export FileCallbackHandler
- added:
export tracing_enabled
- removed:
export BaseCallbackHandler
- removed:
export CHECKMARK_EMOJI - removed:
export EXCEPTION_EMOJI - removed:
export HISTORY_EMOJI - removed:
export THINKING_EMOJI
- added:
export LangChainTracerV1
- added:
export LangChainTracerV1
- added:
export BaseRun - added:
export ChainRun - added:
export LLMRun - added:
export RunTypeEnum - added:
export ToolRun - added:
export TracerSession - added:
export TracerSessionBase - added:
export TracerSessionV1 - added:
export TracerSessionV1Base - added:
export TracerSessionV1Create
- removed:
export PRINT_WARNINGS
- removed:
export APIChain - removed:
export AnalyzeDocumentChain - removed:
export ArangoGraphQAChain - removed:
export ChatVectorDBChain - removed:
export ConstitutionalChain - removed:
export ConversationChain - removed:
export ConversationalRetrievalChain - removed:
export FalkorDBQAChain - removed:
export FlareChain - removed:
export GraphCypherQAChain - removed:
export GraphQAChain - removed:
export GraphSparqlQAChain - removed:
export HugeGraphQAChain - removed:
export HypotheticalDocumentEmbedder - removed:
export KuzuQAChain - removed:
export LLMChain - removed:
export LLMCheckerChain - removed:
export LLMMathChain - removed:
export LLMRequestsChain - removed:
export LLMRouterChain - removed:
export LLMSummarizationCheckerChain - removed:
export MapReduceChain - removed:
export MapReduceDocumentsChain - removed:
export MapRerankDocumentsChain - removed:
export MultiPromptChain - removed:
export MultiRetrievalQAChain - removed:
export MultiRouteChain - removed:
export NatBotChain - removed:
export NebulaGraphQAChain - removed:
export NeptuneOpenCypherQAChain - removed:
export OpenAIModerationChain - removed:
export OpenAPIEndpointChain - removed:
export QAGenerationChain - removed:
export QAWithSourcesChain - removed:
export ReduceDocumentsChain - removed:
export RefineDocumentsChain - removed:
export RetrievalQA - removed:
export RetrievalQAWithSourcesChain - removed:
export RouterChain - removed:
export SequentialChain - removed:
export SimpleSequentialChain - removed:
export StuffDocumentsChain - removed:
export TransformChain - removed:
export VectorDBQA - removed:
export VectorDBQAWithSourcesChain - removed:
export create_citation_fuzzy_match_chain - removed:
export create_extraction_chain - removed:
export create_extraction_chain_pydantic - removed:
export create_history_aware_retriever - removed:
export create_qa_with_sources_chain - removed:
export create_qa_with_structure_chain - removed:
export create_retrieval_chain - removed:
export create_sql_query_chain - removed:
export create_tagging_chain - removed:
export create_tagging_chain_pydantic - removed:
export generate_example - removed:
export load_chain - added:
export create_importer - added:
variable importer
- added:
class APIChain - added:
constructor APIChain(*args: Any, **kwargs: Any) -> None
- added:
async method Chain.aprep_inputs(inputs: Union[Dict[str, Any], Any]) -> Dict[str, str] - added:
async method Chain.aprep_outputs(inputs: Dict[str, str], outputs: Dict[str, str], return_only_outputs: bool=…) -> Dict[str, str]
- added:
field ReduceDocumentsChain.collapse_max_retries: Optional[int]
- added:
class GremlinQAChain(Chain) - added:
field GremlinQAChain.graph: GremlinGraph - added:
field GremlinQAChain.gremlin_fix_chain: LLMChain - added:
field GremlinQAChain.gremlin_generation_chain: LLMChain - added:
field GremlinQAChain.input_key: str - added:
field GremlinQAChain.max_fix_retries: int - added:
field GremlinQAChain.output_key: str - added:
field GremlinQAChain.qa_chain: LLMChain - added:
field GremlinQAChain.return_direct: bool - added:
field GremlinQAChain.return_intermediate_steps: bool - added:
field GremlinQAChain.top_k: int - added:
function extract_gremlin(text: str) -> str - added:
method GremlinQAChain.execute_query(query: str) -> List[Any] - added:
method GremlinQAChain.execute_with_retry(_run_manager: CallbackManagerForChainRun, callbacks: CallbackManager, generated_gremlin: str) -> List[Any] - added:
method GremlinQAChain.log_invalid_query(_run_manager: CallbackManagerForChainRun, generated_query: str, error_message: str) -> None - added:
property GremlinQAChain.input_keys -> List[str] - added:
property GremlinQAChain.output_keys -> List[str] - added:
static method GremlinQAChain.from_llm(llm: BaseLanguageModel, *, gremlin_fix_prompt: BasePromptTemplate=…, qa_prompt: BasePromptTemplate=…, gremlin_prompt: BasePromptTemplate=…, **kwargs: Any) -> GremlinQAChain - added:
variable INTERMEDIATE_STEPS_KEY
- added:
function extract_cypher(text: str) -> str - added:
function remove_prefix(text: str, prefix: str) -> str
- changed:
field NeptuneOpenCypherQAChain.graph: NeptuneGraph → field NeptuneOpenCypherQAChain.graph: BaseNeptuneGraph
- added:
class NeptuneSparqlQAChain(Chain) - added:
field NeptuneSparqlQAChain.extra_instructions: Optional[str] - added:
field NeptuneSparqlQAChain.graph: NeptuneRdfGraph - added:
field NeptuneSparqlQAChain.input_key: str - added:
field NeptuneSparqlQAChain.output_key: str - added:
field NeptuneSparqlQAChain.qa_chain: LLMChain - added:
field NeptuneSparqlQAChain.return_direct: bool - added:
field NeptuneSparqlQAChain.return_intermediate_steps: bool - added:
field NeptuneSparqlQAChain.sparql_generation_chain: LLMChain - added:
field NeptuneSparqlQAChain.top_k: int - added:
function extract_sparql(query: str) -> str - added:
property NeptuneSparqlQAChain.input_keys -> List[str] - added:
property NeptuneSparqlQAChain.output_keys -> List[str] - added:
static method NeptuneSparqlQAChain.from_llm(llm: BaseLanguageModel, *, qa_prompt: BasePromptTemplate=…, sparql_prompt: BasePromptTemplate=…, examples: Optional[str]=…, **kwargs: Any) -> NeptuneSparqlQAChain - added:
variable INTERMEDIATE_STEPS_KEY - added:
variable SPARQL_GENERATION_PROMPT - added:
variable SPARQL_GENERATION_TEMPLATE
- added:
class OntotextGraphDBQAChain(Chain) - added:
field OntotextGraphDBQAChain.graph: OntotextGraphDBGraph - added:
field OntotextGraphDBQAChain.input_key: str - added:
field OntotextGraphDBQAChain.max_fix_retries: int - added:
field OntotextGraphDBQAChain.output_key: str - added:
field OntotextGraphDBQAChain.qa_chain: LLMChain - added:
field OntotextGraphDBQAChain.sparql_fix_chain: LLMChain - added:
field OntotextGraphDBQAChain.sparql_generation_chain: LLMChain - added:
property OntotextGraphDBQAChain.input_keys -> List[str] - added:
property OntotextGraphDBQAChain.output_keys -> List[str] - added:
static method OntotextGraphDBQAChain.from_llm(llm: BaseLanguageModel, *, sparql_generation_prompt: BasePromptTemplate=…, sparql_fix_prompt: BasePromptTemplate=…, max_fix_retries: int=…, qa_prompt: BasePromptTemplate=…, **kwargs: Any) -> OntotextGraphDBQAChain
- added:
variable GRAPHDB_QA_PROMPT - added:
variable GRAPHDB_QA_TEMPLATE - added:
variable GRAPHDB_SPARQL_FIX_PROMPT - added:
variable GRAPHDB_SPARQL_FIX_TEMPLATE - added:
variable GRAPHDB_SPARQL_GENERATION_PROMPT - added:
variable GRAPHDB_SPARQL_GENERATION_TEMPLATE
- added:
field GraphSparqlQAChain.return_sparql_query: bool - added:
field GraphSparqlQAChain.sparql_query_key: str
- added:
function load_llm(*args: Any, **kwargs: Any) -> None - added:
function load_llm_from_config(*args: Any, **kwargs: Any) -> None
- added:
field OpenAIModerationChain.async_client: Any
- removed:
function convert_python_function_to_openai_function(function: Callable) -> Dict[str, Any] - removed:
function convert_to_openai_function(function: Union[Dict[str, Any], Type[BaseModel], Callable]) -> Dict[str, Any] - removed:
function create_openai_fn_runnable(functions: Sequence[Union[Dict[str, Any], Type[BaseModel], Callable]], llm: Runnable, prompt: BasePromptTemplate, *, enforce_single_function_usage: bool=…, output_parser: Optional[Union[BaseOutputParser, BaseGenerationOutputParser]]=…, **kwargs: Any) -> Runnable - removed:
function create_structured_output_runnable(output_schema: Union[Dict[str, Any], Type[BaseModel]], llm: Runnable, prompt: BasePromptTemplate, *, output_parser: Optional[Union[BaseOutputParser, BaseGenerationOutputParser]]=…, **kwargs: Any) -> Runnable - removed:
function get_openai_output_parser(functions: Sequence[Union[Dict[str, Any], Type[BaseModel], Callable]]) -> Union[BaseOutputParser, BaseGenerationOutputParser] - removed:
variable PYTHON_TO_JSON_TYPES - added:
export PYTHON_TO_JSON_TYPES - added:
export convert_to_openai_function - added:
export create_openai_fn_runnable - added:
export create_structured_output_runnable - added:
export get_openai_output_parser
- added:
export load_query_constructor_runnable
- removed:
class Comparator(str, Enum) - removed:
class Comparison(FilterDirective) - removed:
class Expr(BaseModel) - removed:
class FilterDirective(Expr, ABC) - removed:
class Operation(FilterDirective) - removed:
class Operator(str, Enum) - removed:
class StructuredQuery(Expr) - removed:
class Visitor(ABC) - removed:
field Comparator.CONTAIN - removed:
field Comparator.EQ - removed:
field Comparator.GT - removed:
field Comparator.GTE - removed:
field Comparator.IN - removed:
field Comparator.LIKE - removed:
field Comparator.LT - removed:
field Comparator.LTE - removed:
field Comparator.NE - removed:
field Comparator.NIN - removed:
field Comparison.attribute: str - removed:
field Comparison.comparator: Comparator - removed:
field Comparison.value: Any - removed:
field Operation.arguments: List[FilterDirective] - removed:
field Operation.operator: Operator - removed:
field Operator.AND - removed:
field Operator.NOT - removed:
field Operator.OR - removed:
field StructuredQuery.filter: Optional[FilterDirective] - removed:
field StructuredQuery.limit: Optional[int] - removed:
field StructuredQuery.query: str - removed:
field Visitor.allowed_comparators: Optional[Sequence[Comparator]] - removed:
field Visitor.allowed_operators: Optional[Sequence[Operator]] - removed:
method Expr.accept(visitor: Visitor) -> Any - removed:
method Visitor.visit_comparison(comparison: Comparison) -> Any - removed:
method Visitor.visit_operation(operation: Operation) -> Any - removed:
method Visitor.visit_structured_query(structured_query: StructuredQuery) -> Any - added:
export Comparator - added:
export Comparison - added:
export Expr - added:
export FilterDirective - added:
export Operation - added:
export Operator - added:
export StructuredQuery - added:
export Visitor
- removed:
class LoadingCallable(Protocol) - removed:
export BaseCallbackManager - removed:
export BaseCombineDocumentsChain - removed:
export Callbacks - removed:
export LLMChain - removed:
export MAP_RERANK_PROMPT - removed:
export MapReduceDocumentsChain - removed:
export MapRerankDocumentsChain - removed:
export ReduceDocumentsChain - removed:
export RefineDocumentsChain - removed:
export StuffDocumentsChain - removed:
export map_reduce_prompt - removed:
export refine_prompts - removed:
export stuff_prompt - removed:
function load_qa_chain(llm: BaseLanguageModel, chain_type: str=…, verbose: Optional[bool]=…, callback_manager: Optional[BaseCallbackManager]=…, **kwargs: Any) -> BaseCombineDocumentsChain - added:
export LoadingCallable - added:
export load_qa_chain
- added:
class LoadingCallable(Protocol) - added:
function load_qa_chain(llm: BaseLanguageModel, chain_type: str=…, verbose: Optional[bool]=…, callback_manager: Optional[BaseCallbackManager]=…, **kwargs: Any) -> BaseCombineDocumentsChain
- added:
async static method EmbeddingRouterChain.afrom_names_and_descriptions(names_and_descriptions: Sequence[Tuple[str, Sequence[str]]], vectorstore_cls: Type[VectorStore], embeddings: Embeddings, **kwargs: Any) -> EmbeddingRouterChain
- changed:
function create_sql_query_chain(llm: BaseLanguageModel, db: SQLDatabase, prompt: Optional[BasePromptTemplate]=…, k: int=…) -> Runnable[Union[SQLInput, SQLInputWithTables], str] → function create_sql_query_chain(llm: BaseLanguageModel, db: SQLDatabase, prompt: Optional[BasePromptTemplate]=…, k: int=…) -> Runnable[Union[SQLInput, SQLInputWithTables, Dict[str, Any]], str]
- added:
export create_openai_fn_runnable - added:
export create_structured_output_runnable
- added:
function create_openai_fn_runnable(functions: Sequence[Union[Dict[str, Any], Type[BaseModel], Callable]], llm: Runnable, prompt: Optional[BasePromptTemplate]=…, *, enforce_single_function_usage: bool=…, output_parser: Optional[Union[BaseOutputParser, BaseGenerationOutputParser]]=…, **llm_kwargs: Any) -> Runnable - added:
function create_structured_output_runnable(output_schema: Union[Dict[str, Any], Type[BaseModel]], llm: Runnable, prompt: Optional[BasePromptTemplate]=…, *, output_parser: Optional[Union[BaseOutputParser, BaseGenerationOutputParser]]=…, enforce_function_usage: bool=…, return_single: bool=…, mode: Literal['openai-functions', 'openai-tools', 'openai-json']=…, **kwargs: Any) -> Runnable - added:
function get_openai_output_parser(functions: Sequence[Union[Dict[str, Any], Type[BaseModel], Callable]]) -> Union[BaseOutputParser, BaseGenerationOutputParser]
- removed:
class LoadingCallable(Protocol) - removed:
export BaseCombineDocumentsChain - removed:
export Callbacks - removed:
export LLMChain - removed:
export MapReduceDocumentsChain - removed:
export ReduceDocumentsChain - removed:
export RefineDocumentsChain - removed:
export StuffDocumentsChain - removed:
export map_reduce_prompt - removed:
export refine_prompts - removed:
export stuff_prompt - removed:
function load_summarize_chain(llm: BaseLanguageModel, chain_type: str=…, verbose: Optional[bool]=…, **kwargs: Any) -> BaseCombineDocumentsChain - added:
export LoadingCallable - added:
export load_summarize_chain
- added:
class LoadingCallable(Protocol) - added:
function load_summarize_chain(llm: BaseLanguageModel, chain_type: str=…, verbose: Optional[bool]=…, **kwargs: Any) -> BaseCombineDocumentsChain
- removed:
export AzureChatOpenAI - removed:
export BedrockChat - removed:
export ChatAnthropic - removed:
export ChatAnyscale - removed:
export ChatBaichuan - removed:
export ChatCohere - removed:
export ChatDatabricks - removed:
export ChatEverlyAI - removed:
export ChatFireworks - removed:
export ChatGooglePalm - removed:
export ChatHunyuan - removed:
export ChatJavelinAIGateway - removed:
export ChatKonko - removed:
export ChatLiteLLM - removed:
export ChatMLflowAIGateway - removed:
export ChatMlflow - removed:
export ChatOllama - removed:
export ChatOpenAI - removed:
export ChatVertexAI - removed:
export ChatYandexGPT - removed:
export ErnieBotChat - removed:
export FakeListChatModel - removed:
export GigaChat - removed:
export HumanInputChatModel - removed:
export JinaChat - removed:
export MiniMaxChat - removed:
export PaiEasChatEndpoint - removed:
export PromptLayerChatOpenAI - removed:
export QianfanChatEndpoint - removed:
export VolcEngineMaasChat
- removed:
variable PagedPDFSplitter - removed:
variable TelegramChatLoader - added:
export GithubFileLoader - added:
export PagedPDFSplitter - added:
export TelegramChatLoader - added:
export YuqueLoader
- removed:
export RecordHandler
- removed:
export PathLike
- removed:
export PyPDFLoader - added:
export PagedPDFSplitter
- added:
export OpenVINOEmbeddings
- changed:
constructor CacheBackedEmbeddings(underlying_embeddings: Embeddings, document_embedding_store: BaseStore[str, List[float]]) -> None → constructor CacheBackedEmbeddings(underlying_embeddings: Embeddings, document_embedding_store: BaseStore[str, List[float]], *, batch_size: Optional[int]=…) -> None - changed:
static method CacheBackedEmbeddings.from_bytes_store(underlying_embeddings: Embeddings, document_embedding_cache: ByteStore, *, namespace: str=…) -> CacheBackedEmbeddings → static method CacheBackedEmbeddings.from_bytes_store(underlying_embeddings: Embeddings, document_embedding_cache: ByteStore, *, namespace: str=…, batch_size: Optional[int]=…) -> CacheBackedEmbeddings - added:
async method CacheBackedEmbeddings.aembed_documents(texts: List[str]) -> List[List[float]] - added:
async method CacheBackedEmbeddings.aembed_query(text: str) -> List[float]
- removed:
export KG_TRIPLE_DELIMITER
- removed:
async method RecordManager.acreate_schema() -> None - removed:
async method RecordManager.adelete_keys(keys: Sequence[str]) -> None - removed:
async method RecordManager.aexists(keys: Sequence[str]) -> List[bool] - removed:
async method RecordManager.aget_time() -> float - removed:
async method RecordManager.alist_keys(*, before: Optional[float]=…, after: Optional[float]=…, group_ids: Optional[Sequence[str]]=…, limit: Optional[int]=…) -> List[str] - removed:
async method RecordManager.aupdate(keys: Sequence[str], *, group_ids: Optional[Sequence[Optional[str]]]=…, time_at_least: Optional[float]=…) -> None - removed:
class RecordManager(ABC) - removed:
constructor RecordManager(namespace: str) -> None - removed:
method RecordManager.create_schema() -> None - removed:
method RecordManager.delete_keys(keys: Sequence[str]) -> None - removed:
method RecordManager.exists(keys: Sequence[str]) -> List[bool] - removed:
method RecordManager.get_time() -> float - removed:
method RecordManager.list_keys(*, before: Optional[float]=…, after: Optional[float]=…, group_ids: Optional[Sequence[str]]=…, limit: Optional[int]=…) -> List[str] - removed:
method RecordManager.update(keys: Sequence[str], *, group_ids: Optional[Sequence[Optional[str]]]=…, time_at_least: Optional[float]=…) -> None - removed:
variable NAMESPACE_UUID
- removed:
async method GraphIndexCreator.afrom_text(text: str, prompt: BasePromptTemplate=…) -> NetworkxEntityGraph - removed:
class GraphIndexCreator(BaseModel) - removed:
field GraphIndexCreator.graph_type: Type[NetworkxEntityGraph] - removed:
field GraphIndexCreator.llm: Optional[BaseLanguageModel] - removed:
method GraphIndexCreator.from_text(text: str, prompt: BasePromptTemplate=…) -> NetworkxEntityGraph - added:
export GraphIndexCreator - added:
export NetworkxEntityGraph
- added:
variable KG_TRIPLE_DELIMITER
- added:
async method VectorStoreIndexWrapper.aquery(question: str, llm: Optional[BaseLanguageModel]=…, retriever_kwargs: Optional[Dict[str, Any]]=…, **kwargs: Any) -> str - added:
async method VectorStoreIndexWrapper.aquery_with_sources(question: str, llm: Optional[BaseLanguageModel]=…, retriever_kwargs: Optional[Dict[str, Any]]=…, **kwargs: Any) -> dict - added:
async method VectorstoreIndexCreator.afrom_documents(documents: List[Document]) -> VectorStoreIndexWrapper - added:
async method VectorstoreIndexCreator.afrom_loaders(loaders: List[BaseLoader]) -> VectorStoreIndexWrapper
- removed:
export LlamaContentFormatter - added:
export CustomOpenAIContentFormatter
- added:
async method ConversationBufferMemory.abuffer() -> Any - added:
async method ConversationBufferMemory.abuffer_as_messages() -> List[BaseMessage] - added:
async method ConversationBufferMemory.abuffer_as_str() -> str - added:
async method ConversationBufferMemory.aload_memory_variables(inputs: Dict[str, Any]) -> Dict[str, Any] - added:
async method ConversationStringBufferMemory.aclear() -> None - added:
async method ConversationStringBufferMemory.aload_memory_variables(inputs: Dict[str, Any]) -> Dict[str, str] - added:
async method ConversationStringBufferMemory.asave_context(inputs: Dict[str, Any], outputs: Dict[str, str]) -> None
- added:
async method BaseChatMemory.aclear() -> None - added:
async method BaseChatMemory.asave_context(inputs: Dict[str, Any], outputs: Dict[str, str]) -> None
- added:
class SQLiteEntityStore.Config - added:
field SQLiteEntityStore.Config.arbitrary_types_allowed - added:
field SQLiteEntityStore.conn: Any
- removed:
class ConversationKGMemory(BaseChatMemory) - removed:
field ConversationKGMemory.ai_prefix: str - removed:
field ConversationKGMemory.entity_extraction_prompt: BasePromptTemplate - removed:
field ConversationKGMemory.human_prefix: str - removed:
field ConversationKGMemory.k: int - removed:
field ConversationKGMemory.kg: NetworkxEntityGraph - removed:
field ConversationKGMemory.knowledge_extraction_prompt: BasePromptTemplate - removed:
field ConversationKGMemory.llm: BaseLanguageModel - removed:
field ConversationKGMemory.memory_key: str - removed:
field ConversationKGMemory.summary_message_cls: Type[BaseMessage] - removed:
method ConversationKGMemory.clear() -> None - removed:
method ConversationKGMemory.get_current_entities(input_string: str) -> List[str] - removed:
method ConversationKGMemory.get_knowledge_triplets(input_string: str) -> List[KnowledgeTriple] - removed:
method ConversationKGMemory.load_memory_variables(inputs: Dict[str, Any]) -> Dict[str, Any] - removed:
method ConversationKGMemory.save_context(inputs: Dict[str, Any], outputs: Dict[str, str]) -> None - removed:
property ConversationKGMemory.memory_variables -> List[str] - added:
export ConversationKGMemory
- removed:
async method MotorheadMemory.init() -> None - removed:
class MotorheadMemory(BaseChatMemory) - removed:
field MotorheadMemory.api_key: Optional[str] - removed:
field MotorheadMemory.client_id: Optional[str] - removed:
field MotorheadMemory.context: Optional[str] - removed:
field MotorheadMemory.memory_key: str - removed:
field MotorheadMemory.session_id: str - removed:
field MotorheadMemory.timeout: int - removed:
field MotorheadMemory.url: str - removed:
method MotorheadMemory.delete_session() -> None - removed:
method MotorheadMemory.load_memory_variables(values: Dict[str, Any]) -> Dict[str, Any] - removed:
method MotorheadMemory.save_context(inputs: Dict[str, Any], outputs: Dict[str, str]) -> None - removed:
property MotorheadMemory.memory_variables -> List[str] - removed:
variable MANAGED_URL - added:
export MotorheadMemory
- removed:
property ConversationSummaryBufferMemory.buffer -> List[BaseMessage] - added:
property ConversationSummaryBufferMemory.buffer -> Union[str, List[BaseMessage]]
- added:
async method VectorStoreRetrieverMemory.aclear() -> None - added:
async method VectorStoreRetrieverMemory.aload_memory_variables(inputs: Dict[str, Any]) -> Dict[str, Union[List[Document], str]] - added:
async method VectorStoreRetrieverMemory.asave_context(inputs: Dict[str, Any], outputs: Dict[str, str]) -> None
- removed:
class ZepMemory(ConversationBufferMemory) - removed:
constructor ZepMemory(session_id: str, url: str=…, api_key: Optional[str]=…, output_key: Optional[str]=…, input_key: Optional[str]=…, return_messages: bool=…, human_prefix: str=…, ai_prefix: str=…, memory_key: str=…) - removed:
field ZepMemory.chat_memory: ZepChatMessageHistory - removed:
method ZepMemory.save_context(inputs: Dict[str, Any], outputs: Dict[str, str], metadata: Optional[Dict[str, Any]]=…) -> None - added:
export ZepMemory
- removed:
class JsonKeyOutputFunctionsParser(JsonOutputFunctionsParser) - removed:
class JsonOutputFunctionsParser(BaseCumulativeTransformOutputParser[Any]) - removed:
class OutputFunctionsParser(BaseGenerationOutputParser[Any]) - removed:
class PydanticAttrOutputFunctionsParser(PydanticOutputFunctionsParser) - removed:
class PydanticOutputFunctionsParser(OutputFunctionsParser) - removed:
field JsonKeyOutputFunctionsParser.key_name: str - removed:
field JsonOutputFunctionsParser.args_only: bool - removed:
field JsonOutputFunctionsParser.strict: bool - removed:
field OutputFunctionsParser.args_only: bool - removed:
field PydanticAttrOutputFunctionsParser.attr_name: str - removed:
field PydanticOutputFunctionsParser.pydantic_schema: Union[Type[BaseModel], Dict[str, Type[BaseModel]]] - removed:
method JsonKeyOutputFunctionsParser.parse_result(result: List[Generation], *, partial: bool=…) -> Any - removed:
method JsonOutputFunctionsParser.parse(text: str) -> Any - removed:
method JsonOutputFunctionsParser.parse_result(result: List[Generation], *, partial: bool=…) -> Any - removed:
method OutputFunctionsParser.parse_result(result: List[Generation], *, partial: bool=…) -> Any - removed:
method PydanticAttrOutputFunctionsParser.parse_result(result: List[Generation], *, partial: bool=…) -> Any - removed:
method PydanticOutputFunctionsParser.parse_result(result: List[Generation], *, partial: bool=…) -> Any - removed:
method PydanticOutputFunctionsParser.validate_schema(values: Dict) -> Dict - added:
export JsonKeyOutputFunctionsParser - added:
export JsonOutputFunctionsParser - added:
export OutputFunctionsParser - added:
export PydanticAttrOutputFunctionsParser - added:
export PydanticOutputFunctionsParser
- removed:
class JsonKeyOutputFunctionsParser(JsonOutputFunctionsParser) - removed:
class JsonOutputFunctionsParser(BaseCumulativeTransformOutputParser[Any]) - removed:
class OutputFunctionsParser(BaseGenerationOutputParser[Any]) - removed:
class PydanticAttrOutputFunctionsParser(PydanticOutputFunctionsParser) - removed:
class PydanticOutputFunctionsParser(OutputFunctionsParser) - removed:
field JsonKeyOutputFunctionsParser.key_name: str - removed:
field JsonOutputFunctionsParser.args_only: bool - removed:
field JsonOutputFunctionsParser.strict: bool - removed:
field OutputFunctionsParser.args_only: bool - removed:
field PydanticAttrOutputFunctionsParser.attr_name: str - removed:
field PydanticOutputFunctionsParser.pydantic_schema: Union[Type[BaseModel], Dict[str, Type[BaseModel]]] - removed:
method JsonKeyOutputFunctionsParser.parse_result(result: List[Generation], *, partial: bool=…) -> Any - removed:
method JsonOutputFunctionsParser.parse(text: str) -> Any - removed:
method JsonOutputFunctionsParser.parse_result(result: List[Generation], *, partial: bool=…) -> Any - removed:
method OutputFunctionsParser.parse_result(result: List[Generation], *, partial: bool=…) -> Any - removed:
method PydanticAttrOutputFunctionsParser.parse_result(result: List[Generation], *, partial: bool=…) -> Any - removed:
method PydanticOutputFunctionsParser.parse_result(result: List[Generation], *, partial: bool=…) -> Any - removed:
method PydanticOutputFunctionsParser.validate_schema(values: Dict) -> Dict - added:
export JsonKeyOutputFunctionsParser - added:
export JsonOutputFunctionsParser - added:
export PydanticAttrOutputFunctionsParser - added:
export PydanticOutputFunctionsParser
- removed:
class JsonOutputKeyToolsParser(JsonOutputToolsParser) - removed:
class JsonOutputToolsParser(BaseGenerationOutputParser[Any]) - removed:
class PydanticToolsParser(JsonOutputToolsParser) - removed:
field JsonOutputKeyToolsParser.key_name: str - removed:
field PydanticToolsParser.tools: List[Type[BaseModel]] - removed:
method JsonOutputKeyToolsParser.parse_result(result: List[Generation], *, partial: bool=…) -> Any - removed:
method JsonOutputToolsParser.parse_result(result: List[Generation], *, partial: bool=…) -> Any - removed:
method PydanticToolsParser.parse_result(result: List[Generation], *, partial: bool=…) -> Any - added:
export JsonOutputKeyToolsParser - added:
export JsonOutputToolsParser - added:
export PydanticToolsParser
- removed:
class PydanticOutputParser(BaseOutputParser[T]) - removed:
field PydanticOutputParser.pydantic_object: Type[T] - removed:
method PydanticOutputParser.get_format_instructions() -> str - removed:
method PydanticOutputParser.parse(text: str) -> T - removed:
variable T - added:
export PydanticOutputParser
- removed:
class GuardrailsOutputParser(BaseOutputParser) - removed:
field GuardrailsOutputParser.api: Optional[Callable] - removed:
field GuardrailsOutputParser.args: Any - removed:
field GuardrailsOutputParser.guard: Any - removed:
field GuardrailsOutputParser.kwargs: Any - removed:
method GuardrailsOutputParser.get_format_instructions() -> str - removed:
method GuardrailsOutputParser.parse(text: str) -> Dict - removed:
static method GuardrailsOutputParser.from_pydantic(output_class: Any, num_reasks: int=…, api: Optional[Callable]=…, *args: Any, **kwargs: Any) -> GuardrailsOutputParser - removed:
static method GuardrailsOutputParser.from_rail(rail_file: str, num_reasks: int=…, api: Optional[Callable]=…, *args: Any, **kwargs: Any) -> GuardrailsOutputParser - removed:
static method GuardrailsOutputParser.from_rail_string(rail_str: str, num_reasks: int=…, api: Optional[Callable]=…, *args: Any, **kwargs: Any) -> GuardrailsOutputParser - added:
export GuardrailsOutputParser
- removed:
class NGramOverlapExampleSelector(BaseExampleSelector, BaseModel) - removed:
field NGramOverlapExampleSelector.example_prompt: PromptTemplate - removed:
field NGramOverlapExampleSelector.examples: List[dict] - removed:
field NGramOverlapExampleSelector.threshold: float - removed:
function ngram_overlap_score(source: List[str], example: List[str]) -> float - removed:
method NGramOverlapExampleSelector.add_example(example: Dict[str, str]) -> None - removed:
method NGramOverlapExampleSelector.check_dependencies(values: Dict) -> Dict - removed:
method NGramOverlapExampleSelector.select_examples(input_variables: Dict[str, str]) -> List[dict] - added:
export NGramOverlapExampleSelector - added:
export ngram_overlap_score
- added:
export AzureAISearchRetriever - added:
export DriaRetriever - added:
export NeuralDBRetriever
- added:
export AzureAISearchRetriever - added:
export AzureCognitiveSearchRetriever
- removed:
export AzureCognitiveSearchRetriever
- added:
export CrossEncoderReranker - added:
export FlashrankRerank
- removed:
async method BaseDocumentCompressor.acompress_documents(documents: Sequence[Document], query: str, callbacks: Optional[Callbacks]=…) -> Sequence[Document] - removed:
class BaseDocumentCompressor(BaseModel, ABC) - removed:
method BaseDocumentCompressor.compress_documents(documents: Sequence[Document], query: str, callbacks: Optional[Callbacks]=…) -> Sequence[Document]
- changed:
field CohereRerank.client: Client → field CohereRerank.client: Any - changed:
field CohereRerank.top_n: int → field CohereRerank.top_n: Optional[int] - added:
method CohereRerank.rerank(documents: Sequence[Union[str, Document, dict]], query: str, *, model: Optional[str]=…, top_n: Optional[int]=…, max_chunks_per_doc: Optional[int]=…) -> List[Dict[str, Any]]
- added:
class CrossEncoderReranker(BaseDocumentCompressor) - added:
class CrossEncoderReranker.Config - added:
field CrossEncoderReranker.Config.arbitrary_types_allowed - added:
field CrossEncoderReranker.Config.extra - added:
field CrossEncoderReranker.model: BaseCrossEncoder - added:
field CrossEncoderReranker.top_n: int - added:
method CrossEncoderReranker.compress_documents(documents: Sequence[Document], query: str, callbacks: Optional[Callbacks]=…) -> Sequence[Document]
- added:
class FlashrankRerank(BaseDocumentCompressor) - added:
class FlashrankRerank.Config - added:
field FlashrankRerank.Config.arbitrary_types_allowed - added:
field FlashrankRerank.Config.extra - added:
field FlashrankRerank.client: Ranker - added:
field FlashrankRerank.model: Optional[str] - added:
field FlashrankRerank.top_n: int - added:
method FlashrankRerank.compress_documents(documents: Sequence[Document], query: str, callbacks: Optional[Callbacks]=…) -> Sequence[Document] - added:
method FlashrankRerank.validate_environment(values: Dict) -> Dict - added:
variable DEFAULT_MODEL_NAME
- changed:
async method EnsembleRetriever.arank_fusion(query: str, run_manager: AsyncCallbackManagerForRetrieverRun) -> List[Document] → async method EnsembleRetriever.arank_fusion(query: str, run_manager: AsyncCallbackManagerForRetrieverRun, *, config: Optional[RunnableConfig]=…) -> List[Document] - changed:
field EnsembleRetriever.retrievers: List[BaseRetriever] → field EnsembleRetriever.retrievers: List[RetrieverLike] - changed:
method EnsembleRetriever.rank_fusion(query: str, run_manager: CallbackManagerForRetrieverRun) -> List[Document] → method EnsembleRetriever.rank_fusion(query: str, run_manager: CallbackManagerForRetrieverRun, *, config: Optional[RunnableConfig]=…) -> List[Document] - added:
async method EnsembleRetriever.ainvoke(input: str, config: Optional[RunnableConfig]=…, **kwargs: Any) -> List[Document] - added:
function unique_by_key(iterable: Iterable[T], key: Callable[[T], H]) -> Iterator[T] - added:
method EnsembleRetriever.invoke(input: str, config: Optional[RunnableConfig]=…, **kwargs: Any) -> List[Document] - added:
property EnsembleRetriever.config_specs -> List[ConfigurableFieldSpec] - added:
variable H - added:
variable T
- removed:
export DocumentAttributeValueType
- removed:
class LineList(BaseModel) - removed:
constructor LineListOutputParser() -> None - removed:
field LineList.lines: List[str] - changed:
class LineListOutputParser(PydanticOutputParser) → class LineListOutputParser(BaseOutputParser[List[str]]) - changed:
method LineListOutputParser.parse(text: str) -> LineList → method LineListOutputParser.parse(text: str) -> List[str] - changed:
static method MultiQueryRetriever.from_llm(retriever: BaseRetriever, llm: BaseLLM, prompt: PromptTemplate=…, parser_key: str=…, include_original: bool=…) -> MultiQueryRetriever → static method MultiQueryRetriever.from_llm(retriever: BaseRetriever, llm: BaseLanguageModel, prompt: PromptTemplate=…, parser_key: Optional[str]=…, include_original: bool=…) -> MultiQueryRetriever
- added:
field ParentDocumentRetriever.child_metadata_fields: Optional[Sequence[str]]
- added:
class AstraDBTranslator(Visitor) - added:
field AstraDBTranslator.allowed_comparators - added:
field AstraDBTranslator.allowed_operators - added:
method AstraDBTranslator.visit_comparison(comparison: Comparison) -> Dict - added:
method AstraDBTranslator.visit_operation(operation: Operation) -> Dict - added:
method AstraDBTranslator.visit_structured_query(structured_query: StructuredQuery) -> Tuple[str, dict] - added:
variable MULTIPLE_ARITY_COMPARATORS
- added:
variable QUERY_CONSTRUCTOR_RUN_NAME
- added:
class DatabricksVectorSearchTranslator(Visitor) - added:
field DatabricksVectorSearchTranslator.allowed_comparators - added:
field DatabricksVectorSearchTranslator.allowed_operators - added:
method DatabricksVectorSearchTranslator.visit_comparison(comparison: Comparison) -> Dict - added:
method DatabricksVectorSearchTranslator.visit_operation(operation: Operation) -> Dict - added:
method DatabricksVectorSearchTranslator.visit_structured_query(structured_query: StructuredQuery) -> Tuple[str, dict]
- added:
class DingoDBTranslator(Visitor) - added:
field DingoDBTranslator.allowed_comparators - added:
field DingoDBTranslator.allowed_operators - added:
method DingoDBTranslator.visit_comparison(comparison: Comparison) -> Comparison - added:
method DingoDBTranslator.visit_operation(operation: Operation) -> Operation - added:
method DingoDBTranslator.visit_structured_query(structured_query: StructuredQuery) -> Tuple[str, dict]
- changed:
function process_value(value: Union[int, float, str]) -> str → function process_value(value: Union[int, float, str], comparator: Comparator) -> str
- added:
class PGVectorTranslator(Visitor) - added:
field PGVectorTranslator.allowed_comparators - added:
field PGVectorTranslator.allowed_operators - added:
method PGVectorTranslator.visit_comparison(comparison: Comparison) -> Dict - added:
method PGVectorTranslator.visit_operation(operation: Operation) -> Dict - added:
method PGVectorTranslator.visit_structured_query(structured_query: StructuredQuery) -> Tuple[str, dict]
- added:
class TencentVectorDBTranslator(Visitor) - added:
constructor TencentVectorDBTranslator(meta_keys: Optional[Sequence[str]]=…) - added:
field TencentVectorDBTranslator.COMPARATOR_MAP - added:
field TencentVectorDBTranslator.allowed_comparators: Optional[Sequence[Comparator]] - added:
field TencentVectorDBTranslator.allowed_operators: Optional[Sequence[Operator]] - added:
method TencentVectorDBTranslator.visit_comparison(comparison: Comparison) -> str - added:
method TencentVectorDBTranslator.visit_operation(operation: Operation) -> str - added:
method TencentVectorDBTranslator.visit_structured_query(structured_query: StructuredQuery) -> Tuple[str, dict]
- added:
async method TimeWeightedVectorStoreRetriever.aget_salient_docs(query: str) -> Dict[int, Tuple[Document, float]]
- removed:
class LineList(BaseModel) - removed:
constructor QuestionListOutputParser() -> None - removed:
field LineList.lines: List[str] - changed:
class QuestionListOutputParser(PydanticOutputParser) → class QuestionListOutputParser(BaseOutputParser[List[str]]) - changed:
method QuestionListOutputParser.parse(text: str) -> LineList → method QuestionListOutputParser.parse(text: str) -> List[str]
- added:
export tracing_enabled
- added:
export LangChainTracerV1 - added:
export get_headers
- added:
export BaseRun - added:
export ChainRun - added:
export LLMRun - added:
export RunTypeEnum - added:
export ToolRun - added:
export TracerSession - added:
export TracerSessionBase - added:
export TracerSessionV1 - added:
export TracerSessionV1Base - added:
export TracerSessionV1Create
- changed:
field RunEvalConfig.custom_evaluators: Optional[List[Union[RunEvaluator, StringEvaluator]]] → field RunEvalConfig.custom_evaluators: Optional[List[CUSTOM_EVALUATOR_TYPE]] - changed:
field RunEvalConfig.evaluators: List[Union[EvaluatorType, str, EvalConfig]] → field RunEvalConfig.evaluators: List[Union[SINGLE_EVAL_CONFIG_TYPE, CUSTOM_EVALUATOR_TYPE]] - added:
field RunEvalConfig.batch_evaluators: Optional[List[BATCH_EVALUATOR_LIKE]] - added:
variable BATCH_EVALUATOR_LIKE - added:
variable CUSTOM_EVALUATOR_TYPE - added:
variable RUN_EVALUATOR_LIKE - added:
variable SINGLE_EVAL_CONFIG_TYPE
- changed:
async function arun_on_dataset(client: Optional[Client], dataset_name: str, llm_or_chain_factory: MODEL_OR_CHAIN_FACTORY, *, evaluation: Optional[smith_eval.RunEvalConfig]=…, concurrency_level: int=…, project_name: Optional[str]=…, project_metadata: Optional[Dict[str, Any]]=…, verbose: bool=…, tags: Optional[List[str]]=…, **kwargs: Any) -> Dict[str, Any] → async function arun_on_dataset(client: Optional[Client], dataset_name: str, llm_or_chain_factory: MODEL_OR_CHAIN_FACTORY, *, evaluation: Optional[smith_eval.RunEvalConfig]=…, dataset_version: Optional[Union[datetime, str]]=…, concurrency_level: int=…, project_name: Optional[str]=…, project_metadata: Optional[Dict[str, Any]]=…, verbose: bool=…, revision_id: Optional[str]=…, **kwargs: Any) -> Dict[… - changed:
function run_on_dataset(client: Optional[Client], dataset_name: str, llm_or_chain_factory: MODEL_OR_CHAIN_FACTORY, *, evaluation: Optional[smith_eval.RunEvalConfig]=…, concurrency_level: int=…, project_name: Optional[str]=…, project_metadata: Optional[Dict[str, Any]]=…, verbose: bool=…, tags: Optional[List[str]]=…, **kwargs: Any) -> Dict[str, Any] → function run_on_dataset(client: Optional[Client], dataset_name: str, llm_or_chain_factory: MODEL_OR_CHAIN_FACTORY, *, evaluation: Optional[smith_eval.RunEvalConfig]=…, dataset_version: Optional[Union[datetime, str]]=…, concurrency_level: int=…, project_name: Optional[str]=…, project_metadata: Optional[Dict[str, Any]]=…, verbose: bool=…, revision_id: Optional[str]=…, **kwargs: Any) -> Dict[str, An… - added:
class ChatModelInput(TypedDict) - added:
field ChatModelInput.messages: List[BaseMessage]
- added:
export InvalidKeyException
- added:
async method EncoderBackedStore.amdelete(keys: Sequence[K]) -> None - added:
async method EncoderBackedStore.amget(keys: Sequence[K]) -> List[Optional[V]] - added:
async method EncoderBackedStore.amset(key_value_pairs: Sequence[Tuple[K, V]]) -> None - added:
async method EncoderBackedStore.ayield_keys(*, prefix: Optional[str]=…) -> Union[AsyncIterator[K], AsyncIterator[str]]
- changed:
constructor LocalFileStore(root_path: Union[str, Path]) -> None → constructor LocalFileStore(root_path: Union[str, Path], *, chmod_file: Optional[int]=…, chmod_dir: Optional[int]=…, update_atime: bool=…) -> None
- removed:
class InMemoryBaseStore(BaseStore[str, V], Generic[V]) - removed:
constructor InMemoryBaseStore() -> None - removed:
method InMemoryBaseStore.mdelete(keys: Sequence[str]) -> None - removed:
method InMemoryBaseStore.mget(keys: Sequence[str]) -> List[Optional[V]] - removed:
method InMemoryBaseStore.mset(key_value_pairs: Sequence[Tuple[str, V]]) -> None - removed:
method InMemoryBaseStore.yield_keys(prefix: Optional[str]=…) -> Iterator[str] - removed:
variable InMemoryByteStore - removed:
variable InMemoryStore - removed:
variable V - added:
export InMemoryBaseStore - added:
export InMemoryByteStore - added:
export InMemoryStore
- removed:
class CharacterTextSplitter(TextSplitter) - removed:
class ElementType(TypedDict) - removed:
class HTMLHeaderTextSplitter - removed:
class HeaderType(TypedDict) - removed:
class Language(str, Enum) - removed:
class LatexTextSplitter(RecursiveCharacterTextSplitter) - removed:
class LineType(TypedDict) - removed:
class MarkdownHeaderTextSplitter - removed:
class MarkdownTextSplitter(RecursiveCharacterTextSplitter) - removed:
class NLTKTextSplitter(TextSplitter) - removed:
class PythonCodeTextSplitter(RecursiveCharacterTextSplitter) - removed:
class RecursiveCharacterTextSplitter(TextSplitter) - removed:
class SentenceTransformersTokenTextSplitter(TextSplitter) - removed:
class SpacyTextSplitter(TextSplitter) - removed:
class TextSplitter(BaseDocumentTransformer, ABC) - removed:
class TokenTextSplitter(TextSplitter) - removed:
class Tokenizer - removed:
constructor CharacterTextSplitter(separator: str=…, is_separator_regex: bool=…, **kwargs: Any) -> None - removed:
constructor HTMLHeaderTextSplitter(headers_to_split_on: List[Tuple[str, str]], return_each_element: bool=…) - removed:
constructor LatexTextSplitter(**kwargs: Any) -> None - removed:
constructor MarkdownHeaderTextSplitter(headers_to_split_on: List[Tuple[str, str]], return_each_line: bool=…, strip_headers: bool=…) - removed:
constructor MarkdownTextSplitter(**kwargs: Any) -> None - removed:
constructor NLTKTextSplitter(separator: str=…, language: str=…, **kwargs: Any) -> None - removed:
constructor PythonCodeTextSplitter(**kwargs: Any) -> None - removed:
constructor RecursiveCharacterTextSplitter(separators: Optional[List[str]]=…, keep_separator: bool=…, is_separator_regex: bool=…, **kwargs: Any) -> None - removed:
constructor SentenceTransformersTokenTextSplitter(chunk_overlap: int=…, model_name: str=…, tokens_per_chunk: Optional[int]=…, **kwargs: Any) -> None - removed:
constructor SpacyTextSplitter(separator: str=…, pipeline: str=…, max_length: int=…, **kwargs: Any) -> None - removed:
constructor TextSplitter(chunk_size: int=…, chunk_overlap: int=…, length_function: Callable[[str], int]=…, keep_separator: bool=…, add_start_index: bool=…, strip_whitespace: bool=…) -> None - removed:
constructor TokenTextSplitter(encoding_name: str=…, model_name: Optional[str]=…, allowed_special: Union[Literal['all'], AbstractSet[str]]=…, disallowed_special: Union[Literal['all'], Collection[str]]=…, **kwargs: Any) -> None - removed:
field ElementType.content: str - removed:
field ElementType.metadata: Dict[str, str] - removed:
field ElementType.url: str - removed:
field ElementType.xpath: str - removed:
field HeaderType.data: str - removed:
field HeaderType.level: int - removed:
field HeaderType.name: str - removed:
field Language.COBOL - removed:
field Language.CPP - removed:
field Language.CSHARP - removed:
field Language.GO - removed:
field Language.HTML - removed:
field Language.JAVA - removed:
field Language.JS - removed:
field Language.KOTLIN - removed:
field Language.LATEX - removed:
field Language.MARKDOWN - removed:
field Language.PHP - removed:
field Language.PROTO - removed:
field Language.PYTHON - removed:
field Language.RST - removed:
field Language.RUBY - removed:
field Language.RUST - removed:
field Language.SCALA - removed:
field Language.SOL - removed:
field Language.SWIFT - removed:
field Language.TS - removed:
field LineType.content: str - removed:
field LineType.metadata: Dict[str, str] - removed:
field Tokenizer.chunk_overlap: int - removed:
field Tokenizer.decode: Callable[[List[int]], str] - removed:
field Tokenizer.encode: Callable[[str], List[int]] - removed:
field Tokenizer.tokens_per_chunk: int - removed:
function split_text_on_tokens(*, text: str, tokenizer: Tokenizer) -> List[str] - removed:
method CharacterTextSplitter.split_text(text: str) -> List[str] - removed:
method HTMLHeaderTextSplitter.aggregate_elements_to_chunks(elements: List[ElementType]) -> List[Document] - removed:
method HTMLHeaderTextSplitter.split_text(text: str) -> List[Document] - removed:
method HTMLHeaderTextSplitter.split_text_from_file(file: Any) -> List[Document] - removed:
method HTMLHeaderTextSplitter.split_text_from_url(url: str) -> List[Document] - removed:
method MarkdownHeaderTextSplitter.aggregate_lines_to_chunks(lines: List[LineType]) -> List[Document] - removed:
method MarkdownHeaderTextSplitter.split_text(text: str) -> List[Document] - removed:
method NLTKTextSplitter.split_text(text: str) -> List[str] - removed:
method RecursiveCharacterTextSplitter.split_text(text: str) -> List[str] - removed:
method SentenceTransformersTokenTextSplitter.count_tokens(*, text: str) -> int - removed:
method SentenceTransformersTokenTextSplitter.split_text(text: str) -> List[str] - removed:
method SpacyTextSplitter.split_text(text: str) -> List[str] - removed:
method TextSplitter.create_documents(texts: List[str], metadatas: Optional[List[dict]]=…) -> List[Document] - removed:
method TextSplitter.split_documents(documents: Iterable[Document]) -> List[Document] - removed:
method TextSplitter.split_text(text: str) -> List[str] - removed:
method TextSplitter.transform_documents(documents: Sequence[Document], **kwargs: Any) -> Sequence[Document] - removed:
method TokenTextSplitter.split_text(text: str) -> List[str] - removed:
static method RecursiveCharacterTextSplitter.from_language(language: Language, **kwargs: Any) -> RecursiveCharacterTextSplitter - removed:
static method RecursiveCharacterTextSplitter.get_separators_for_language(language: Language) -> List[str] - removed:
static method TextSplitter.from_huggingface_tokenizer(tokenizer: Any, **kwargs: Any) -> TextSplitter - removed:
static method TextSplitter.from_tiktoken_encoder(encoding_name: str=…, model_name: Optional[str]=…, allowed_special: Union[Literal['all'], AbstractSet[str]]=…, disallowed_special: Union[Literal['all'], Collection[str]]=…, **kwargs: Any) -> TS - removed:
variable TS - removed:
variable logger - added:
export CharacterTextSplitter - added:
export ElementType - added:
export HTMLHeaderTextSplitter - added:
export HeaderType - added:
export KonlpyTextSplitter - added:
export Language - added:
export LatexTextSplitter - added:
export LineType - added:
export MarkdownHeaderTextSplitter - added:
export MarkdownTextSplitter - added:
export NLTKTextSplitter - added:
export PythonCodeTextSplitter - added:
export RecursiveCharacterTextSplitter - added:
export RecursiveJsonSplitter - added:
export SentenceTransformersTokenTextSplitter - added:
export SpacyTextSplitter - added:
export TextSplitter - added:
export TokenTextSplitter - added:
export Tokenizer - added:
export split_text_on_tokens
- removed:
function render_text_description(tools: List[BaseTool]) -> str - removed:
function render_text_description_and_args(tools: List[BaseTool]) -> str - added:
export ToolsRenderer - added:
export render_text_description - added:
export render_text_description_and_args
- removed:
class RetrieverInput(BaseModel) - removed:
field RetrieverInput.query: str - removed:
function create_retriever_tool(retriever: BaseRetriever, name: str, description: str) -> Tool - added:
export RetrieverInput - added:
export ToolsRenderer - added:
export create_retriever_tool - added:
export render_text_description - added:
export render_text_description_and_args
- added:
export AlphaVantageAPIWrapper - added:
export ApifyWrapper - added:
export ArceeWrapper - added:
export ArxivAPIWrapper - added:
export BibtexparserWrapper - added:
export BingSearchAPIWrapper - added:
export BraveSearchWrapper - added:
export DuckDuckGoSearchAPIWrapper - added:
export GoldenQueryAPIWrapper - added:
export GoogleFinanceAPIWrapper - added:
export GoogleJobsAPIWrapper - added:
export GoogleLensAPIWrapper - added:
export GooglePlacesAPIWrapper - added:
export GoogleScholarAPIWrapper - added:
export GoogleSearchAPIWrapper - added:
export GoogleSerperAPIWrapper - added:
export GoogleTrendsAPIWrapper - added:
export GraphQLAPIWrapper - added:
export JiraAPIWrapper - added:
export LambdaWrapper - added:
export MaxComputeAPIWrapper - added:
export MerriamWebsterAPIWrapper - added:
export MetaphorSearchAPIWrapper - added:
export NasaAPIWrapper - added:
export OpenWeatherMapAPIWrapper - added:
export OutlineAPIWrapper - added:
export Portkey - added:
export PowerBIDataset - added:
export PubMedAPIWrapper - added:
export PythonREPL - added:
export SQLDatabase - added:
export SceneXplainAPIWrapper - added:
export SearchApiAPIWrapper - added:
export SearxSearchWrapper - added:
export SerpAPIWrapper - added:
export SparkSQL - added:
export StackExchangeAPIWrapper - added:
export SteamWebAPIWrapper - added:
export TensorflowDatasets - added:
export TwilioAPIWrapper - added:
export WikipediaAPIWrapper - added:
export WolframAlphaAPIWrapper - added:
export ZapierNLAWrapper
- removed:
export TextRequestsWrapper
- removed:
export Matrix
- added:
export AlibabaCloudOpenSearch - added:
export AlibabaCloudOpenSearchSettings - added:
export AnalyticDB - added:
export Annoy - added:
export AstraDB - added:
export AtlasDB - added:
export AwaDB - added:
export AzureCosmosDBVectorSearch - added:
export AzureSearch - added:
export Bagel - added:
export Cassandra - added:
export Chroma - added:
export Clarifai - added:
export Clickhouse - added:
export ClickhouseSettings - added:
export DashVector - added:
export DatabricksVectorSearch - added:
export DeepLake - added:
export Dingo - added:
export DocArrayHnswSearch - added:
export DocArrayInMemorySearch - added:
export DuckDB - added:
export EcloudESVectorStore - added:
export ElasticKnnSearch - added:
export ElasticVectorSearch - added:
export ElasticsearchStore - added:
export Epsilla - added:
export FAISS - added:
export Hologres - added:
export LLMRails - added:
export LanceDB - added:
export Marqo - added:
export MatchingEngine - added:
export Meilisearch - added:
export Milvus - added:
export MomentoVectorIndex - added:
export MongoDBAtlasVectorSearch - added:
export MyScale - added:
export MyScaleSettings - added:
export Neo4jVector - added:
export NeuralDBVectorStore - added:
export OpenSearchVectorSearch - added:
export PGEmbedding - added:
export PGVector - added:
export Pinecone - added:
export Qdrant - added:
export Redis - added:
export Rockset - added:
export SKLearnVectorStore - added:
export SQLiteVSS - added:
export ScaNN - added:
export SemaDB - added:
export SingleStoreDB - added:
export StarRocks - added:
export SupabaseVectorStore - added:
export Tair - added:
export TencentVectorDB - added:
export Tigris - added:
export TileDB - added:
export TimescaleVector - added:
export Typesense - added:
export USearch - added:
export Vald - added:
export Vearch - added:
export Vectara - added:
export VespaStore - added:
export Weaviate - added:
export Yellowbrick - added:
export ZepVectorStore - added:
export Zilliz
- removed:
export CosmosDBDocumentType
- removed:
export CVST
- removed:
export MongoDBDocumentType
- removed:
export SingleStoreDBRetriever