langchain[minor]: Deprecate LLMChain, ConversationalRetrievalQAChain,…

… RetrievalQAChain (langchain-ai#5225)

* langchain[minor]: Deprecate LLMChain, ConversationalRetrievalQAChain, RetrievalQAChain

* chore: lint files

* slightly betta docstrings

* chore: lint files

* update 0.2 to 0.3

langchain-core/src/language_models/chat_models.ts

@@ -446,8 +446,6 @@ export abstract class BaseChatModel<
   }
 
   /**
-   * @deprecated Use .invoke() instead. Will be removed in 0.2.0.
-   *
    * Generates chat based on the input messages.
    * @param messages An array of arrays of BaseMessage instances.
    * @param options The call options or an array of stop sequences.

langchain-core/src/language_models/llms.ts

@@ -402,8 +402,6 @@ export abstract class BaseLLM<
   }
 
   /**
-   * @deprecated Use .invoke() instead. Will be removed in 0.2.0.
-   *
    * Run the LLM on the given prompts and input, handling caching.
    */
   async generate(

langchain-core/src/retrievers/index.ts

@@ -77,7 +77,7 @@ export abstract class BaseRetriever<
   }
 
   /**
-   * @deprecated Use .invoke() instead. Will be removed in 0.2.0.
+   * @deprecated Use .invoke() instead. Will be removed in 0.3.0.
    *
    * Main method used to retrieve relevant documents. It takes a query
    * string and an optional configuration object, and returns a promise that

langchain-core/src/tools.ts

@@ -43,7 +43,7 @@ export interface StructuredToolInterface<
   schema: T | z.ZodEffects<T>;
 
   /**
-   * @deprecated Use .invoke() instead. Will be removed in 0.2.0.
+   * @deprecated Use .invoke() instead. Will be removed in 0.3.0.
    *
    * Calls the tool with the provided argument, configuration, and tags. It
    * parses the input according to the schema, handles any errors, and
@@ -107,7 +107,7 @@ export abstract class StructuredTool<
   }
 
   /**
-   * @deprecated Use .invoke() instead. Will be removed in 0.2.0.
+   * @deprecated Use .invoke() instead. Will be removed in 0.3.0.
    *
    * Calls the tool with the provided argument, configuration, and tags. It
    * parses the input according to the schema, handles any errors, and
@@ -172,7 +172,7 @@ export abstract class StructuredTool<
 
 export interface ToolInterface extends StructuredToolInterface {
   /**
-   * @deprecated Use .invoke() instead. Will be removed in 0.2.0.
+   * @deprecated Use .invoke() instead. Will be removed in 0.3.0.
    *
    * Calls the tool with the provided argument and callbacks. It handles
    * string inputs specifically.
@@ -199,7 +199,7 @@ export abstract class Tool extends StructuredTool {
   }
 
   /**
-   * @deprecated Use .invoke() instead. Will be removed in 0.2.0.
+   * @deprecated Use .invoke() instead. Will be removed in 0.3.0.
    *
    * Calls the tool with the provided argument and callbacks. It handles
    * string inputs specifically.
@@ -273,7 +273,7 @@ export class DynamicTool extends Tool {
   }
 
   /**
-   * @deprecated Use .invoke() instead. Will be removed in 0.2.0.
+   * @deprecated Use .invoke() instead. Will be removed in 0.3.0.
    */
   async call(
     arg: string | undefined | z.input<this["schema"]>,
@@ -328,7 +328,7 @@ export class DynamicStructuredTool<
   }
 
   /**
-   * @deprecated Use .invoke() instead. Will be removed in 0.2.0.
+   * @deprecated Use .invoke() instead. Will be removed in 0.3.0.
    */
   async call(
     arg: z.output<T>,

langchain/src/chains/conversational_retrieval_chain.ts

@@ -33,37 +33,79 @@ export interface ConversationalRetrievalQAChainInput extends ChainInputs {
 }
 
 /**
+ * @deprecated This class will be removed in 0.3.0. See below for an example implementation using
+ * `createRetrievalChain`.
+ *
  * Class for conducting conversational question-answering tasks with a
  * retrieval component. Extends the BaseChain class and implements the
  * ConversationalRetrievalQAChainInput interface.
  * @example
  * ```typescript
- * const model = new ChatAnthropic({});
- *
- * const text = fs.readFileSync("state_of_the_union.txt", "utf8");
- *
- * const textSplitter = new RecursiveCharacterTextSplitter({ chunkSize: 1000 });
- * const docs = await textSplitter.createDocuments([text]);
+ * import { ChatAnthropic } from "@langchain/anthropic";
+ * import {
+ *   ChatPromptTemplate,
+ *   MessagesPlaceholder,
+ * } from "@langchain/core/prompts";
+ * import { BaseMessage } from "@langchain/core/messages";
+ * import { createStuffDocumentsChain } from "langchain/chains/combine_documents";
+ * import { createHistoryAwareRetriever } from "langchain/chains/history_aware_retriever";
+ * import { createRetrievalChain } from "langchain/chains/retrieval";
  *
- * const vectorStore = await HNSWLib.fromDocuments(docs, new OpenAIEmbeddings());
+ * const retriever = ...your retriever;
+ * const llm = new ChatAnthropic();
  *
- * const chain = ConversationalRetrievalQAChain.fromLLM(
- *   model,
- *   vectorStore.asRetriever(),
- * );
+ * // Contextualize question
+ * const contextualizeQSystemPrompt = `
+ * Given a chat history and the latest user question
+ * which might reference context in the chat history,
+ * formulate a standalone question which can be understood
+ * without the chat history. Do NOT answer the question, just
+ * reformulate it if needed and otherwise return it as is.`;
+ * const contextualizeQPrompt = ChatPromptTemplate.fromMessages([
+ *   ["system", contextualizeQSystemPrompt],
+ *   new MessagesPlaceholder("chat_history"),
+ *   ["human", "{input}"],
+ * ]);
+ * const historyAwareRetriever = await createHistoryAwareRetriever({
+ *   llm,
+ *   retriever,
+ *   rephrasePrompt: contextualizeQPrompt,
+ * });
  *
- * const question = "What did the president say about Justice Breyer?";
+ * // Answer question
+ * const qaSystemPrompt = `
+ * You are an assistant for question-answering tasks. Use
+ * the following pieces of retrieved context to answer the
+ * question. If you don't know the answer, just say that you
+ * don't know. Use three sentences maximum and keep the answer
+ * concise.
+ * \n\n
+ * {context}`;
+ * const qaPrompt = ChatPromptTemplate.fromMessages([
+ *   ["system", qaSystemPrompt],
+ *   new MessagesPlaceholder("chat_history"),
+ *   ["human", "{input}"],
+ * ]);
  *
- * const res = await chain.call({ question, chat_history: "" });
- * console.log(res);
+ * // Below we use createStuffDocuments_chain to feed all retrieved context
+ * // into the LLM. Note that we can also use StuffDocumentsChain and other
+ * // instances of BaseCombineDocumentsChain.
+ * const questionAnswerChain = await createStuffDocumentsChain({
+ *   llm,
+ *   prompt: qaPrompt,
+ * });
  *
- * const chatHistory = `${question}\n${res.text}`;
- * const followUpRes = await chain.call({
- *   question: "Was that nice?",
- *   chat_history: chatHistory,
+ * const ragChain = await createRetrievalChain({
+ *   retriever: historyAwareRetriever,
+ *   combineDocsChain: questionAnswerChain,
  * });
- * console.log(followUpRes);
  *
+ * // Usage:
+ * const chat_history: BaseMessage[] = [];
+ * const response = await ragChain.invoke({
+ *   chat_history,
+ *   input: "...",
+ * });
  * ```
  */
 export class ConversationalRetrievalQAChain

langchain/src/chains/llm_chain.ts

@@ -72,16 +72,21 @@ function _getLanguageModel(llmLike: RunnableInterface): BaseLanguageModel {
 }
 
 /**
+ * @deprecated This class will be removed in 0.3.0. Use the LangChain Expression Language (LCEL) instead.
+ * See the example below for how to use LCEL with the LLMChain class:
+ *
  * Chain to run queries against LLMs.
  *
  * @example
  * ```ts
- * import { LLMChain } from "langchain/chains";
- * import { OpenAI } from "langchain/llms/openai";
- * import { PromptTemplate } from "langchain/prompts";
+ * import { ChatPromptTemplate } from "@langchain/core/prompts";
+ * import { ChatOpenAI } from "@langchain/openai";
+ *
+ * const prompt = ChatPromptTemplate.fromTemplate("Tell me a {adjective} joke");
+ * const llm = new ChatOpenAI();
+ * const chain = prompt.pipe(llm);
  *
- * const prompt = PromptTemplate.fromTemplate("Tell me a {adjective} joke");
- * const llm = new LLMChain({ llm: new OpenAI(), prompt });
+ * const response = await chain.invoke({ adjective: "funny" });
  * ```
  */
 export class LLMChain<

langchain/src/chains/retrieval_qa.ts

@@ -23,28 +23,37 @@ export interface RetrievalQAChainInput extends Omit<ChainInputs, "memory"> {
 }
 
 /**
+ * @deprecated This class will be removed in 0.3.0. See below for an example implementation using
+ * `createRetrievalChain`:
  * Class representing a chain for performing question-answering tasks with
  * a retrieval component.
  * @example
  * ```typescript
- * // Initialize the OpenAI model and the remote retriever with the specified configuration
- * const model = new ChatOpenAI({});
- * const retriever = new RemoteLangChainRetriever({
- *   url: "http://example.com/api",
- *   auth: { bearer: "foo" },
- *   inputKey: "message",
- *   responseKey: "response",
- * });
+ * import { createStuffDocumentsChain } from "langchain/chains/combine_documents";
+ * import { ChatPromptTemplate } from "@langchain/core/prompts";
+ * import { createRetrievalChain } from "langchain/chains/retrieval";
+ * import { MemoryVectorStore } from "langchain/vectorstores/memory";
+ *
+ * const documents = [...your documents here];
+ * const embeddings = ...your embeddings model;
+ * const llm = ...your LLM model;
  *
- * // Create a RetrievalQAChain using the model and retriever
- * const chain = RetrievalQAChain.fromLLM(model, retriever);
+ * const vectorstore = await MemoryVectorStore.fromDocuments(
+ *   documents,
+ *   embeddings
+ * );
+ * const prompt = ChatPromptTemplate.fromTemplate(`Answer the user's question: {input}`);
  *
- * // Execute the chain with a query and log the result
- * const res = await chain.call({
- *   query: "What did the president say about Justice Breyer?",
+ * const combineDocsChain = await createStuffDocumentsChain({
+ *   llm,
+ *   prompt,
  * });
- * console.log({ res });
+ * const retriever = vectorstore.asRetriever();
  *
+ * const retrievalChain = await createRetrievalChain({
+ *   combineDocsChain,
+ *   retriever,
+ * });
  * ```
  */
 export class RetrievalQAChain