Merge branch 'main' into brace/tools-return-tool-message

.github/workflows/standard-tests.yml

@@ -29,6 +29,7 @@ jobs:
           GOOGLE_API_KEY: ${{ secrets.GOOGLE_API_KEY }}
           GROQ_API_KEY: ${{ secrets.GROQ_API_KEY }}
           MISTRAL_API_KEY: ${{ secrets.MISTRAL_API_KEY }}
+          DISABLE_CONSOLE_LOGS: "true"
 
   # The `@langchain/openai` package contains standard tests for ChatOpenAI and AzureChatOpenAI
   # We want to run these separately, so we need to pass the exact path for each test, which means
@@ -50,6 +51,7 @@ jobs:
         run: yarn workspace @langchain/openai test:single src/tests/chat_models_structured_output.int.test.ts src/tests/chat_models-extended.int.test.ts src/tests/chat_models-vision.int.test.ts src/tests/chat_models.int.test.ts src/tests/chat_models.standard.int.test.ts
         env:
           OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
+          DISABLE_CONSOLE_LOGS: "true"
 
   standard-tests-azure-openai:
     runs-on: ubuntu-latest
@@ -71,6 +73,7 @@ jobs:
           AZURE_OPENAI_API_DEPLOYMENT_NAME: "chat"
           AZURE_OPENAI_API_VERSION: ${{ secrets.AZURE_OPENAI_API_VERSION }}
           AZURE_OPENAI_BASE_PATH: ${{ secrets.AZURE_OPENAI_BASE_PATH }}
+          DISABLE_CONSOLE_LOGS: "true"
 
   standard-tests-bedrock:
     runs-on: ubuntu-latest
@@ -90,4 +93,5 @@ jobs:
         env:
           BEDROCK_AWS_REGION: "us-east-1"
           BEDROCK_AWS_SECRET_ACCESS_KEY: ${{ secrets.BEDROCK_AWS_SECRET_ACCESS_KEY }}
-          BEDROCK_AWS_ACCESS_KEY_ID: ${{ secrets.BEDROCK_AWS_ACCESS_KEY_ID }}
+          BEDROCK_AWS_ACCESS_KEY_ID: ${{ secrets.BEDROCK_AWS_ACCESS_KEY_ID }}
+          DISABLE_CONSOLE_LOGS: "true"

.github/workflows/unit-tests-integrations.yml

@@ -11,8 +11,7 @@ on:
     paths:
       - 'langchain-core/**'
       - 'libs/**/**'
-  workflow_dispatch:  # Allows triggering the workflow manually in GitHub UI
-
+  workflow_dispatch: # Allows triggering the workflow manually in GitHub UI
 
 # If another push to the same PR or branch happens while this workflow is still running,
 # cancel the earlier run in favor of the next run.
@@ -48,6 +47,7 @@ jobs:
       PACKAGES: "anthropic,azure-openai,cloudflare,cohere,core,community,exa,google-common,google-gauth,google-genai,google-vertexai,google-vertexai-web,google-webauth,groq,mistralai,mongo,nomic,openai,pinecone,qdrant,redis,textsplitters,weaviate,yandex,baidu-qianfan"
     outputs:
       matrix: ${{ steps.set-matrix.outputs.matrix }}
+      matrix_length: ${{ steps.set-matrix.outputs.matrix_length }}
     steps:
       - id: set-matrix
         run: |
@@ -71,14 +71,17 @@ jobs:
             fi
           done
           matrix="$matrix]}"
+          matrix_length=$(echo $matrix | jq '.include | length')
           echo "Matrix: $matrix"
+          echo "Matrix length: $matrix_length"
+          echo "matrix_length=$matrix_length" >> $GITHUB_OUTPUT
           echo "matrix=$matrix" >> $GITHUB_OUTPUT
 
   unit-tests:
     name: Unit Tests
     needs: prepare-matrix
     # Only run this job if there are packages to test
-    if: ${{ fromJson(needs.prepare-matrix.outputs.matrix).include.length != 0 }}
+    if: needs.prepare-matrix.outputs.matrix_length > 0
     runs-on: ubuntu-latest
     strategy:
       matrix: ${{ fromJson(needs.prepare-matrix.outputs.matrix) }}

CONTRIBUTING.md

@@ -37,16 +37,15 @@ To make creating packages like this easier, we offer the [`create-langchain-inte
 $ npx create-langchain-integration
 ```
 
-After creating the new integration package, you should add it to the [`unit-tests-integrations.yml`](./.github/workflows/unit-tests-integrations.yml) GitHub action workflow so that it is tested in CI. To do this, simply add the integration name inside the `jobs.unit-tests.strategy.matrix.package` array:
+After creating the new integration package, you should add it to the [`unit-tests-integrations.yml`](./.github/workflows/unit-tests-integrations.yml) GitHub action workflow so that it is tested in CI. To do this,simply update the `env` section of the `prepare-matrix` job with your package name inside the `PACKAGES` variable:
 
 ```yaml
-jobs:
-  unit-tests:
-    name: Unit Tests
-    strategy:
-      matrix:
-        package: [anthropic, azure-openai, cloudflare, <your package name>]
-        ...
+prepare-matrix:
+  needs: get-changed-files
+  runs-on: ubuntu-latest
+  env:
+    PACKAGES: "anthropic,azure-openai,cloudflare,<your-package>"
+    ...
 ```
 
 ### Want to add a feature that's already in Python?
@@ -159,6 +158,7 @@ cd libs/langchain-community
 ```
 
 ### Setup
+
 **Prerequisite**: Node version 18+ is required. Please check node version `node -v` and update it if required.
 
 To get started, you will need to install the dependencies for the project. To do so, run:

docs/core_docs/docs/concepts.mdx

@@ -434,7 +434,7 @@ An essential component of a conversation is being able to refer to information i
 At bare minimum, a conversational system should be able to access some window of past messages directly.
 
 The concept of `ChatHistory` refers to a class in LangChain which can be used to wrap an arbitrary chain.
-This `ChatHistory` will keep track of inputs and outputs of the underlying chain, and append them as messages to a message database
+This `ChatHistory` will keep track of inputs and outputs of the underlying chain, and append them as messages to a message database.
 Future interactions will then load those messages and pass them into the chain as part of the input.
 
 ### Document

docs/core_docs/docs/integrations/chat/cohere.mdx

@@ -1,16 +1,11 @@
 ---
 sidebar_label: Cohere
-sidebar_class_name: beta
 ---
 
 import CodeBlock from "@theme/CodeBlock";
 
 # ChatCohere
 
-:::info
-The Cohere Chat API is still in beta. This means Cohere may make breaking changes at any time.
-:::
-
 ## Setup
 
 In order to use the LangChain.js Cohere integration you'll need an API key.
@@ -48,6 +43,14 @@ import ChatStreamExample from "@examples/models/chat/cohere/chat_stream_cohere.t
 You can see a LangSmith trace of this example [here](https://smith.langchain.com/public/36ae0564-b096-4ec1-9318-1f82fe705fe8/r)
 :::
 
+### Tools
+
+The Cohere API supports tool calling, along with multi-hop-tool calling. The following example demonstrates how to call tools:
+
+import ToolCallingExample from "@examples/models/chat/cohere/tool_calling.ts";
+
+<CodeBlock language="typescript">{ToolCallingExample}</CodeBlock>
+
 ### Stateful conversation API
 
 Cohere's chat API supports stateful conversations.

docs/core_docs/docs/integrations/chat/index.mdx

@@ -28,7 +28,7 @@ The table shows, for each integration, which features have been implemented with
 | ChatAnthropic           |   ✅   |    ❌     |                     ✅                     |                                              ✅                                              |                      ✅                       |
 | ChatBaiduWenxin         |   ❌   |    ❌     |                     ❌                     |                                              ❌                                              |                      ❌                       |
 | ChatCloudflareWorkersAI |   ✅   |    ❌     |                     ❌                     |                                              ❌                                              |                      ❌                       |
-| ChatCohere              |   ✅   |    ❌     |                     ❌                     |                                              ❌                                              |                      ❌                       |
+| ChatCohere              |   ✅   |    ❌     |                     ✅                     |                                              ✅                                              |                      ❌                       |
 | ChatFireworks           |   ✅   |    ✅     |                     ✅                     |                                              ❌                                              |                      ❌                       |
 | ChatGoogleGenerativeAI  |   ✅   |    ❌     |                     ❌                     |                                              ❌                                              |                      ✅                       |
 | ChatVertexAI            |   ✅   |    ❌     |                     ✅                     |                                              ✅                                              |                      ✅                       |

docs/core_docs/docs/integrations/vectorstores/redis.mdx

@@ -29,7 +29,7 @@ import IntegrationInstallTooltip from "@mdx_components/integration_install_toolt
 <IntegrationInstallTooltip></IntegrationInstallTooltip>
 
 ```bash npm2yarn
-npm install @langchain/openai @langchain/community
+npm install @langchain/openai @langchain/core @langchain/redis langchain
 ```
 
 ## Index docs

docs/core_docs/package.json

@@ -50,7 +50,7 @@
   },
   "devDependencies": {
     "@babel/eslint-parser": "^7.18.2",
-    "@langchain/langgraph": "latest",
+    "@langchain/langgraph": "0.0.26",
     "@langchain/scripts": "workspace:*",
     "@swc/core": "^1.3.62",
     "@types/cookie": "^0",

examples/src/document_compressors/cohere_rerank.ts

@@ -27,7 +27,7 @@ const docs = [
 
 const cohereRerank = new CohereRerank({
   apiKey: process.env.COHERE_API_KEY, // Default
-  model: "rerank-english-v2.0", // Default
+  model: "rerank-english-v2.0",
 });
 
 const rerankedDocuments = await cohereRerank.rerank(docs, query, {

examples/src/document_compressors/cohere_rerank_compressor.ts

@@ -28,7 +28,7 @@ const docs = [
 const cohereRerank = new CohereRerank({
   apiKey: process.env.COHERE_API_KEY, // Default
   topN: 3, // Default
-  model: "rerank-english-v2.0", // Default
+  model: "rerank-english-v2.0",
 });
 
 const rerankedDocuments = await cohereRerank.compressDocuments(docs, query);

examples/src/embeddings/cohere.ts

@@ -1,7 +1,7 @@
 import { CohereEmbeddings } from "@langchain/cohere";
 
 export const run = async () => {
-  const model = new CohereEmbeddings();
+  const model = new CohereEmbeddings({ model: "embed-english-v3.0" });
   const res = await model.embedQuery(
     "What would be a good company name a company that makes colorful socks?"
   );

examples/src/guides/expression_language/runnable_maps_sequence.ts

@@ -12,7 +12,7 @@ import { MemoryVectorStore } from "langchain/vectorstores/memory";
 const model = new ChatAnthropic();
 const vectorstore = await MemoryVectorStore.fromDocuments(
   [{ pageContent: "mitochondria is the powerhouse of the cell", metadata: {} }],
-  new CohereEmbeddings()
+  new CohereEmbeddings({ model: "embed-english-v3.0" })
 );
 const retriever = vectorstore.asRetriever();
 const template = `Answer the question based only on the following context:

examples/src/indexes/vector_stores/mongodb_atlas_fromTexts.ts

@@ -10,7 +10,7 @@ const collection = client.db(dbName).collection(collectionName);
 const vectorstore = await MongoDBAtlasVectorSearch.fromTexts(
   ["Hello world", "Bye bye", "What's this?"],
   [{ id: 2 }, { id: 1 }, { id: 3 }],
-  new CohereEmbeddings(),
+  new CohereEmbeddings({ model: "embed-english-v3.0" }),
   {
     collection,
     indexName: "default", // The name of the Atlas search index. Defaults to "default"

examples/src/indexes/vector_stores/mongodb_atlas_search.ts

@@ -7,12 +7,15 @@ const namespace = "langchain.test";
 const [dbName, collectionName] = namespace.split(".");
 const collection = client.db(dbName).collection(collectionName);
 
-const vectorStore = new MongoDBAtlasVectorSearch(new CohereEmbeddings(), {
-  collection,
-  indexName: "default", // The name of the Atlas search index. Defaults to "default"
-  textKey: "text", // The name of the collection field containing the raw content. Defaults to "text"
-  embeddingKey: "embedding", // The name of the collection field containing the embedded text. Defaults to "embedding"
-});
+const vectorStore = new MongoDBAtlasVectorSearch(
+  new CohereEmbeddings({ model: "embed-english-v3.0" }),
+  {
+    collection,
+    indexName: "default", // The name of the Atlas search index. Defaults to "default"
+    textKey: "text", // The name of the collection field containing the raw content. Defaults to "text"
+    embeddingKey: "embedding", // The name of the collection field containing the embedded text. Defaults to "embedding"
+  }
+);
 
 const resultOne = await vectorStore.similaritySearch("Hello world", 1);
 console.log(resultOne);

examples/src/indexes/vector_stores/mongodb_metadata_filtering.ts

@@ -9,12 +9,15 @@ const namespace = "langchain.test";
 const [dbName, collectionName] = namespace.split(".");
 const collection = client.db(dbName).collection(collectionName);
 
-const vectorStore = new MongoDBAtlasVectorSearch(new CohereEmbeddings(), {
-  collection,
-  indexName: "default", // The name of the Atlas search index. Defaults to "default"
-  textKey: "text", // The name of the collection field containing the raw content. Defaults to "text"
-  embeddingKey: "embedding", // The name of the collection field containing the embedded text. Defaults to "embedding"
-});
+const vectorStore = new MongoDBAtlasVectorSearch(
+  new CohereEmbeddings({ model: "embed-english-v3.0" }),
+  {
+    collection,
+    indexName: "default", // The name of the Atlas search index. Defaults to "default"
+    textKey: "text", // The name of the collection field containing the raw content. Defaults to "text"
+    embeddingKey: "embedding", // The name of the collection field containing the embedded text. Defaults to "embedding"
+  }
+);
 
 await vectorStore.addDocuments([
   {

examples/src/indexes/vector_stores/mongodb_mmr.ts

@@ -7,12 +7,15 @@ const namespace = "langchain.test";
 const [dbName, collectionName] = namespace.split(".");
 const collection = client.db(dbName).collection(collectionName);
 
-const vectorStore = new MongoDBAtlasVectorSearch(new CohereEmbeddings(), {
-  collection,
-  indexName: "default", // The name of the Atlas search index. Defaults to "default"
-  textKey: "text", // The name of the collection field containing the raw content. Defaults to "text"
-  embeddingKey: "embedding", // The name of the collection field containing the embedded text. Defaults to "embedding"
-});
+const vectorStore = new MongoDBAtlasVectorSearch(
+  new CohereEmbeddings({ model: "embed-english-v3.0" }),
+  {
+    collection,
+    indexName: "default", // The name of the Atlas search index. Defaults to "default"
+    textKey: "text", // The name of the collection field containing the raw content. Defaults to "text"
+    embeddingKey: "embedding", // The name of the collection field containing the embedded text. Defaults to "embedding"
+  }
+);
 
 const resultOne = await vectorStore.maxMarginalRelevanceSearch("Hello world", {
   k: 4,

examples/src/indexes/vector_stores/vercel_postgres/example.ts

@@ -16,7 +16,7 @@ const config = {
 };
 
 const vercelPostgresStore = await VercelPostgres.initialize(
-  new CohereEmbeddings(),
+  new CohereEmbeddings({ model: "embed-english-v3.0" }),
   config
 );
 

examples/src/models/chat/cohere/chat_cohere.ts

@@ -3,7 +3,6 @@ import { ChatPromptTemplate } from "@langchain/core/prompts";
 
 const model = new ChatCohere({
   apiKey: process.env.COHERE_API_KEY, // Default
-  model: "command", // Default
 });
 const prompt = ChatPromptTemplate.fromMessages([
   ["ai", "You are a helpful assistant"],
@@ -14,26 +13,28 @@ const response = await chain.invoke({
   input: "Hello there friend!",
 });
 console.log("response", response);
-/**
-response AIMessage {
-  lc_serializable: true,
-  lc_namespace: [ 'langchain_core', 'messages' ],
-  content: "Hi there! I'm not your friend, but I'm happy to help you in whatever way I can today. How are you doing? Is there anything I can assist you with? I am an AI chatbot capable of generating thorough responses, and I'm designed to have helpful, inclusive conversations with users. \n" +
-    '\n' +
-    "If you have any questions, feel free to ask away, and I'll do my best to provide you with helpful responses. \n" +
-    '\n' +
-    'Would you like me to help you with anything in particular right now?',
-  additional_kwargs: {
-    response_id: 'c6baa057-ef94-4bb0-9c25-3a424963a074',
-    generationId: 'd824fcdc-b922-4ae6-8d45-7b65a21cdd6a',
-    token_count: {
-      prompt_tokens: 66,
-      response_tokens: 104,
-      total_tokens: 170,
-      billed_tokens: 159
+/*
+  response AIMessage {
+    content: 'Hello there! How can I help you today?',
+    name: undefined,
+    additional_kwargs: {
+      response_id: '51ff9e7e-7419-43db-a8e6-17db54805695',
+      generationId: 'f9b507f5-5296-40c5-834c-b1c09e24a0f6',
+      chatHistory: [ [Object], [Object], [Object] ],
+      finishReason: 'COMPLETE',
+      meta: { apiVersion: [Object], billedUnits: [Object], tokens: [Object] }
     },
-    meta: { api_version: [Object], billed_units: [Object] },
-    tool_inputs: null
+    response_metadata: {
+      estimatedTokenUsage: { completionTokens: 10, promptTokens: 78, totalTokens: 88 },
+      response_id: '51ff9e7e-7419-43db-a8e6-17db54805695',
+      generationId: 'f9b507f5-5296-40c5-834c-b1c09e24a0f6',
+      chatHistory: [ [Object], [Object], [Object] ],
+      finishReason: 'COMPLETE',
+      meta: { apiVersion: [Object], billedUnits: [Object], tokens: [Object] }
+    },
+    id: undefined,
+    tool_calls: [],
+    invalid_tool_calls: [],
+    usage_metadata: { input_tokens: 78, output_tokens: 10, total_tokens: 88 }
   }
-}
- */
+*/

examples/src/models/chat/cohere/chat_stream_cohere.ts

@@ -4,7 +4,6 @@ import { StringOutputParser } from "@langchain/core/output_parsers";
 
 const model = new ChatCohere({
   apiKey: process.env.COHERE_API_KEY, // Default
-  model: "command", // Default
 });
 const prompt = ChatPromptTemplate.fromMessages([
   ["ai", "You are a helpful assistant"],
@@ -23,15 +22,16 @@ for await (const item of response) {
 }
 console.log("stream tokens:", streamTokens);
 console.log("stream iters:", streamIters);
-/**
-stream item:
-stream item: Hello! I'm here to help answer any questions you
-stream item:  might have or assist you with any task you'd like to
-stream item:  accomplish. I can provide information
-stream item:  on a wide range of topics
-stream item: , from math and science to history and literature. I can
-stream item:  also help you manage your schedule, set reminders, and
-stream item:  much more. Is there something specific you need help with? Let
-stream item:  me know!
-stream item:
- */
+
+/*
+  stream item:
+  stream item: Hello! I'm here to help answer any questions you
+  stream item:  might have or assist you with any task you'd like to
+  stream item:  accomplish. I can provide information
+  stream item:  on a wide range of topics
+  stream item: , from math and science to history and literature. I can
+  stream item:  also help you manage your schedule, set reminders, and
+  stream item:  much more. Is there something specific you need help with? Let
+  stream item:  me know!
+  stream item:
+*/