diff --git a/docs/core_docs/docs/integrations/text_embedding/pinecone.ipynb b/docs/core_docs/docs/integrations/text_embedding/pinecone.ipynb
new file mode 100644
index 000000000000..e3ab07565f27
--- /dev/null
+++ b/docs/core_docs/docs/integrations/text_embedding/pinecone.ipynb
@@ -0,0 +1,344 @@
+{
+ "cells": [
+ {
+ "cell_type": "raw",
+ "id": "afaf8039",
+ "metadata": {
+ "vscode": {
+ "languageId": "raw"
+ }
+ },
+ "source": [
+ "---\n",
+ "sidebar_label: Pinecone\n",
+ "---"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "9a3d6f34",
+ "metadata": {},
+ "source": [
+ "# PineconeEmbeddings\n",
+ "\n",
+ "This will help you get started with PineconeEmbeddings [embedding models](/docs/concepts/embedding_models) using LangChain. For detailed documentation on `PineconeEmbeddings` features and configuration options, please refer to the [API reference](https://api.js.langchain.com/classes/_langchain_pinecone.PineconeEmbeddings.html).\n",
+ "\n",
+ "## Overview\n",
+ "### Integration details\n",
+ "\n",
+ "| Class | Package | Local | [Py support](https://python.langchain.com/docs/integrations/text_embedding/pinecone/) | Package downloads | Package latest |\n",
+ "| :--- | :--- | :---: | :---: | :---: | :---: |\n",
+ "| [PineconeEmbeddings](https://api.js.langchain.com/classes/_langchain_pinecone.PineconeEmbeddings.html) | [@langchain/pinecone](https://api.js.langchain.com/classes/_langchain_pinecone.PineconeEmbeddings.html) | ❌ | ✅ |  |  |\n",
+ "\n",
+ "## Setup\n",
+ "\n",
+ "To access Pinecone embedding models you'll need to create a Pinecone account, get an API key, and install the `@langchain/pinecone` integration package.\n",
+ "\n",
+ "### Credentials\n",
+ "\n",
+ "Sign up for a [Pinecone](https://www.pinecone.io/) account, retrieve your API key, and set it as an environment variable named `PINECONE_API_KEY`:\n",
+ "\n",
+ "```typescript\n",
+ "process.env.PINECONE_API_KEY = \"your-pinecone-api-key\";\n",
+ "```\n",
+ "\n",
+ "If you want to get automated tracing of your model calls you can also set your [LangSmith](https://docs.smith.langchain.com/) API key by uncommenting below:\n",
+ "\n",
+ "```bash\n",
+ "# export LANGCHAIN_TRACING_V2=\"true\"\n",
+ "# export LANGCHAIN_API_KEY=\"your-api-key\"\n",
+ "```\n",
+ "\n",
+ "### Installation\n",
+ "\n",
+ "The LangChain PineconeEmbeddings integration lives in the `@langchain/pinecone` package:\n",
+ "\n",
+ "```{=mdx}\n",
+ "import IntegrationInstallTooltip from \"@mdx_components/integration_install_tooltip.mdx\";\n",
+ "import Npm2Yarn from \"@theme/Npm2Yarn\";\n",
+ "\n",
+ "\n",
+ "\n",
+ "\n",
+ " @langchain/pinecone @langchain/core\n",
+ "\n",
+ "```"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "45dd1724",
+ "metadata": {},
+ "source": [
+ "## Instantiation\n",
+ "\n",
+ "Now we can instantiate our model object and generate chat completions:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "id": "9ea7a09b",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import { PineconeEmbeddings } from \"@langchain/pinecone\";\n",
+ "\n",
+ "const embeddings = new PineconeEmbeddings({\n",
+ " model: \"multilingual-e5-large\",\n",
+ "});"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "77d271b6",
+ "metadata": {},
+ "source": [
+ "## Indexing and Retrieval\n",
+ "\n",
+ "Embedding models are often used in retrieval-augmented generation (RAG) flows, both as part of indexing data as well as later retrieving it. For more detailed instructions, please see our RAG tutorials under the [working with external knowledge tutorials](/docs/tutorials/#working-with-external-knowledge).\n",
+ "\n",
+ "Below, see how to index and retrieve data using the `embeddings` object we initialized above. In this example, we will index and retrieve a sample document using the demo [`MemoryVectorStore`](/docs/integrations/vectorstores/memory)."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "id": "d817716b",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "LangChain is the framework for building context-aware reasoning applications\n"
+ ]
+ }
+ ],
+ "source": [
+ "// Create a vector store with a sample text\n",
+ "import { MemoryVectorStore } from \"langchain/vectorstores/memory\";\n",
+ "\n",
+ "const text = \"LangChain is the framework for building context-aware reasoning applications\";\n",
+ "\n",
+ "const vectorstore = await MemoryVectorStore.fromDocuments(\n",
+ " [{ pageContent: text, metadata: {} }],\n",
+ " embeddings,\n",
+ ");\n",
+ "\n",
+ "// Use the vector store as a retriever that returns a single document\n",
+ "const retriever = vectorstore.asRetriever(1);\n",
+ "\n",
+ "// Retrieve the most similar text\n",
+ "const retrievedDocuments = await retriever.invoke(\"What is LangChain?\");\n",
+ "\n",
+ "retrievedDocuments[0].pageContent;"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "e02b9855",
+ "metadata": {},
+ "source": [
+ "## Direct Usage\n",
+ "\n",
+ "Under the hood, the vectorstore and retriever implementations are calling `embeddings.embedDocument(...)` and `embeddings.embedQuery(...)` to create embeddings for the text(s) used in `fromDocuments` and the retriever's `invoke` operations, respectively.\n",
+ "\n",
+ "You can directly call these methods to get embeddings for your own use cases.\n",
+ "\n",
+ "### Embed single texts\n",
+ "\n",
+ "You can embed queries for search with `embedQuery`. This generates a vector representation specific to the query:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "id": "0d2befcd",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "[\n",
+ " 0.0191650390625, 0.004924774169921875, -0.015838623046875,\n",
+ " -0.04248046875, 0.040191650390625, -0.02679443359375,\n",
+ " -0.0240936279296875, 0.058624267578125, 0.027069091796875,\n",
+ " -0.0435791015625, 0.01934814453125, 0.040191650390625,\n",
+ " -0.0194244384765625, 0.01386260986328125, -0.0216827392578125,\n",
+ " -0.01073455810546875, -0.0166168212890625, 0.01073455810546875,\n",
+ " -0.0228271484375, 0.0062255859375, 0.035064697265625,\n",
+ " -0.0114593505859375, -0.0257110595703125, -0.0285797119140625,\n",
+ " 0.01190185546875, -0.022186279296875, -0.01500701904296875,\n",
+ " -0.03240966796875, 0.0019063949584960938, -0.039337158203125,\n",
+ " -0.0047454833984375, -0.03125, -0.0123291015625,\n",
+ " -0.00899505615234375, -0.02880859375, 0.014678955078125,\n",
+ " 0.0452880859375, 0.05035400390625, -0.053436279296875,\n",
+ " 0.0265960693359375, -0.0206756591796875, 0.06658935546875,\n",
+ " -0.032989501953125, -0.00724029541015625, 0.0024967193603515625,\n",
+ " 0.0282135009765625, 0.047088623046875, -0.0255126953125,\n",
+ " -0.008453369140625, -0.0039215087890625, 0.0282135009765625,\n",
+ " 0.0270843505859375, -0.0133056640625, -0.0296173095703125,\n",
+ " -0.0455322265625, 0.0225982666015625, -0.04803466796875,\n",
+ " -0.00891876220703125, -0.04669189453125, 0.022064208984375,\n",
+ " -0.0266571044921875, -0.01480865478515625, 0.0295257568359375,\n",
+ " -0.01561737060546875, -0.0411376953125, 0.01345062255859375,\n",
+ " 0.0219879150390625, -0.012786865234375, -0.051727294921875,\n",
+ " -0.0002830028533935547, 0.00690460205078125, -0.01303863525390625,\n",
+ " -0.0457763671875, -0.026763916015625, -0.0181121826171875,\n",
+ " 0.00946807861328125, 0.0250244140625, -0.01458740234375,\n",
+ " 0.0394287109375, -0.0162200927734375, 0.05169677734375,\n",
+ " 0.01126861572265625, 0.01265716552734375, -0.009307861328125,\n",
+ " 0.052490234375, 0.0135345458984375, 0.01332855224609375,\n",
+ " 0.040130615234375, 0.0638427734375, 0.0181121826171875,\n",
+ " 0.004207611083984375, 0.0771484375, 0.024078369140625,\n",
+ " 0.012420654296875, -0.030517578125, -0.0019245147705078125,\n",
+ " 0.0243682861328125, 0.0254974365234375, 0.0036334991455078125,\n",
+ " -0.004550933837890625\n",
+ "]\n"
+ ]
+ }
+ ],
+ "source": [
+ "const singleVector = await embeddings.embedQuery(text);\n",
+ "\n",
+ "console.log(singleVector.slice(0, 100));"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "1b5a7d03",
+ "metadata": {},
+ "source": [
+ "### Embed multiple texts\n",
+ "\n",
+ "You can embed multiple texts for indexing with `embedDocuments`. The internals used for this method may (but do not have to) differ from embedding queries:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "id": "2f4d6e97",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "[\n",
+ " 0.0190887451171875, 0.00482940673828125, -0.0158233642578125,\n",
+ " -0.04254150390625, 0.040130615234375, -0.0268096923828125,\n",
+ " -0.02392578125, 0.058624267578125, 0.0269927978515625,\n",
+ " -0.04345703125, 0.0193328857421875, 0.040374755859375,\n",
+ " -0.0196075439453125, 0.01384735107421875, -0.021881103515625,\n",
+ " -0.01068878173828125, -0.016510009765625, 0.01079559326171875,\n",
+ " -0.0227813720703125, 0.00634765625, 0.035064697265625,\n",
+ " -0.0113983154296875, -0.0257720947265625, -0.0285491943359375,\n",
+ " 0.011749267578125, -0.0222625732421875, -0.0148468017578125,\n",
+ " -0.0325927734375, 0.00203704833984375, -0.0394287109375,\n",
+ " -0.004878997802734375, -0.0311126708984375, -0.01248931884765625,\n",
+ " -0.00897979736328125, -0.0286407470703125, 0.0146331787109375,\n",
+ " 0.04522705078125, 0.050201416015625, -0.053314208984375,\n",
+ " 0.0265960693359375, -0.0207366943359375, 0.06658935546875,\n",
+ " -0.03302001953125, -0.0073699951171875, 0.0024261474609375,\n",
+ " 0.028228759765625, 0.04705810546875, -0.0255279541015625,\n",
+ " -0.0084075927734375, -0.003814697265625, 0.0281524658203125,\n",
+ " 0.0272064208984375, -0.01322174072265625, -0.0295257568359375,\n",
+ " -0.045623779296875, 0.022735595703125, -0.0478515625,\n",
+ " -0.00885772705078125, -0.046844482421875, 0.022003173828125,\n",
+ " -0.026458740234375, -0.0148468017578125, 0.0295562744140625,\n",
+ " -0.01555633544921875, -0.041229248046875, 0.01336669921875,\n",
+ " 0.022125244140625, -0.01276397705078125, -0.051666259765625,\n",
+ " -0.0002474784851074219, 0.006740570068359375, -0.01306915283203125,\n",
+ " -0.04583740234375, -0.026611328125, -0.0182342529296875,\n",
+ " 0.00946044921875, 0.0250701904296875, -0.0146942138671875,\n",
+ " 0.039459228515625, -0.016265869140625, 0.051788330078125,\n",
+ " 0.01110076904296875, 0.0126953125, -0.00925445556640625,\n",
+ " 0.052581787109375, 0.01363372802734375, 0.01332855224609375,\n",
+ " 0.04010009765625, 0.0638427734375, 0.018157958984375,\n",
+ " 0.0040740966796875, 0.07720947265625, 0.0240325927734375,\n",
+ " 0.0123443603515625, -0.0302886962890625, -0.001865386962890625,\n",
+ " 0.024383544921875, 0.025604248046875, 0.00353240966796875,\n",
+ " -0.004474639892578125\n",
+ "]\n",
+ "[\n",
+ " 0.0053253173828125, 0.01305389404296875, -0.0253448486328125,\n",
+ " -0.04241943359375, 0.034942626953125, -0.017425537109375,\n",
+ " -0.02783203125, 0.064208984375, 0.0244903564453125,\n",
+ " -0.0467529296875, 0.021209716796875, 0.02191162109375,\n",
+ " -0.03131103515625, -0.019073486328125, -0.01413726806640625,\n",
+ " -0.008636474609375, -0.011627197265625, 0.0229339599609375,\n",
+ " -0.00762939453125, 0.00594329833984375, 0.0201263427734375,\n",
+ " -0.0247802734375, -0.05047607421875, -0.03765869140625,\n",
+ " 0.0034332275390625, -0.014617919921875, -0.043548583984375,\n",
+ " -0.03594970703125, 0.0002884864807128906, -0.03656005859375,\n",
+ " -0.0102691650390625, 0.0121307373046875, -0.0284271240234375,\n",
+ " -0.0113525390625, -0.01195526123046875, 0.01143646240234375,\n",
+ " 0.051727294921875, 0.0230712890625, -0.046417236328125,\n",
+ " 0.0198211669921875, -0.02337646484375, 0.040985107421875,\n",
+ " -0.03314208984375, -0.025909423828125, -0.00809478759765625,\n",
+ " 0.0291595458984375, 0.04296875, -0.016143798828125,\n",
+ " 0.005706787109375, 0.00860595703125, -0.0035343170166015625,\n",
+ " 0.0118560791015625, -0.0135650634765625, -0.0294036865234375,\n",
+ " -0.029876708984375, 0.03515625, -0.0545654296875,\n",
+ " 0.006862640380859375, -0.041839599609375, 0.021148681640625,\n",
+ " -0.0279998779296875, -0.00949859619140625, 0.03314208984375,\n",
+ " -0.002727508544921875, -0.0400390625, 0.01311492919921875,\n",
+ " 0.01177215576171875, -0.0010013580322265625, -0.052001953125,\n",
+ " 0.00112152099609375, -0.00815582275390625, 0.0321044921875,\n",
+ " -0.0496826171875, -0.0151519775390625, -0.0262908935546875,\n",
+ " -0.005207061767578125, 0.0207977294921875, -0.022705078125,\n",
+ " 0.009735107421875, 0.000682830810546875, 0.05792236328125,\n",
+ " -0.0145263671875, 0.03643798828125, 0.0018339157104492188,\n",
+ " 0.047210693359375, 0.0017986297607421875, 0.0300140380859375,\n",
+ " 0.027923583984375, 0.044708251953125, 0.027618408203125,\n",
+ " 0.00104522705078125, 0.05987548828125, 0.06304931640625,\n",
+ " -0.039703369140625, -0.0386962890625, 0.00797271728515625,\n",
+ " 0.0254974365234375, 0.0245819091796875, 0.010467529296875,\n",
+ " -0.0080413818359375\n",
+ "]\n"
+ ]
+ }
+ ],
+ "source": [
+ "const text2 = \"LangGraph is a library for building stateful, multi-actor applications with LLMs\";\n",
+ "\n",
+ "const vectors = await embeddings.embedDocuments([text, text2]);\n",
+ "\n",
+ "console.log(vectors[0].slice(0, 100));\n",
+ "console.log(vectors[1].slice(0, 100));"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "8938e581",
+ "metadata": {},
+ "source": [
+ "## API reference\n",
+ "\n",
+ "For detailed documentation of all PineconeEmbeddings features and configurations head to the API reference: https://api.js.langchain.com/classes/_langchain_pinecone.PineconeEmbeddings.html"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "TypeScript",
+ "language": "typescript",
+ "name": "tslab"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "mode": "typescript",
+ "name": "javascript",
+ "typescript": true
+ },
+ "file_extension": ".ts",
+ "mimetype": "text/typescript",
+ "name": "typescript",
+ "version": "3.7.2"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/examples/src/embeddings/pinecone.ts b/examples/src/embeddings/pinecone.ts
new file mode 100644
index 000000000000..e7613ea81065
--- /dev/null
+++ b/examples/src/embeddings/pinecone.ts
@@ -0,0 +1,12 @@
+import { PineconeEmbeddings } from "@langchain/pinecone";
+
+export const run = async () => {
+ const model = new PineconeEmbeddings();
+ console.log({ model }); // Prints out model metadata
+ const res = await model.embedQuery(
+ "What would be a good company name a company that makes colorful socks?"
+ );
+ console.log({ res });
+};
+
+await run();
diff --git a/examples/src/indexes/vector_stores/pinecone/delete_docs.ts b/examples/src/indexes/vector_stores/pinecone/delete_docs.ts
index 33d325e996a2..48ce230ffa93 100644
--- a/examples/src/indexes/vector_stores/pinecone/delete_docs.ts
+++ b/examples/src/indexes/vector_stores/pinecone/delete_docs.ts
@@ -5,7 +5,7 @@ import { OpenAIEmbeddings } from "@langchain/openai";
import { PineconeStore } from "@langchain/pinecone";
// Instantiate a new Pinecone client, which will automatically read the
-// env vars: PINECONE_API_KEY and PINECONE_ENVIRONMENT which come from
+// env vars: PINECONE_API_KEY which comes from
// the Pinecone dashboard at https://app.pinecone.io
const pinecone = new Pinecone();
diff --git a/examples/src/indexes/vector_stores/pinecone/index_docs.ts b/examples/src/indexes/vector_stores/pinecone/index_docs.ts
index 84bbccf934c5..31d2a584b2d0 100644
--- a/examples/src/indexes/vector_stores/pinecone/index_docs.ts
+++ b/examples/src/indexes/vector_stores/pinecone/index_docs.ts
@@ -3,9 +3,10 @@ import { Pinecone } from "@pinecone-database/pinecone";
import { Document } from "@langchain/core/documents";
import { OpenAIEmbeddings } from "@langchain/openai";
import { PineconeStore } from "@langchain/pinecone";
+// import { Index } from "@upstash/vector";
// Instantiate a new Pinecone client, which will automatically read the
-// env vars: PINECONE_API_KEY and PINECONE_ENVIRONMENT which come from
+// env vars: PINECONE_API_KEY which comes from
// the Pinecone dashboard at https://app.pinecone.io
const pinecone = new Pinecone();
diff --git a/examples/src/indexes/vector_stores/pinecone/mmr.ts b/examples/src/indexes/vector_stores/pinecone/mmr.ts
index 10a85253b3ea..82aafc797b20 100644
--- a/examples/src/indexes/vector_stores/pinecone/mmr.ts
+++ b/examples/src/indexes/vector_stores/pinecone/mmr.ts
@@ -4,7 +4,7 @@ import { OpenAIEmbeddings } from "@langchain/openai";
import { PineconeStore } from "@langchain/pinecone";
// Instantiate a new Pinecone client, which will automatically read the
-// env vars: PINECONE_API_KEY and PINECONE_ENVIRONMENT which come from
+// env vars: PINECONE_API_KEY which comes from
// the Pinecone dashboard at https://app.pinecone.io
const pinecone = new Pinecone();
diff --git a/examples/src/indexes/vector_stores/pinecone/query_docs.ts b/examples/src/indexes/vector_stores/pinecone/query_docs.ts
index 6294067724ec..1df62db7f1ef 100644
--- a/examples/src/indexes/vector_stores/pinecone/query_docs.ts
+++ b/examples/src/indexes/vector_stores/pinecone/query_docs.ts
@@ -4,7 +4,7 @@ import { OpenAIEmbeddings } from "@langchain/openai";
import { PineconeStore } from "@langchain/pinecone";
// Instantiate a new Pinecone client, which will automatically read the
-// env vars: PINECONE_API_KEY and PINECONE_ENVIRONMENT which come from
+// env vars: PINECONE_API_KEY which comes from
// the Pinecone dashboard at https://app.pinecone.io
const pinecone = new Pinecone();
diff --git a/examples/src/retrievers/pinecone_self_query.ts b/examples/src/retrievers/pinecone_self_query.ts
index 2ae7116c44f1..66d7dfafdc96 100644
--- a/examples/src/retrievers/pinecone_self_query.ts
+++ b/examples/src/retrievers/pinecone_self_query.ts
@@ -83,14 +83,8 @@ const attributeInfo: AttributeInfo[] = [
* Next, we instantiate a vector store. This is where we store the embeddings of the documents.
* We also need to provide an embeddings object. This is used to embed the documents.
*/
-if (
- !process.env.PINECONE_API_KEY ||
- !process.env.PINECONE_ENVIRONMENT ||
- !process.env.PINECONE_INDEX
-) {
- throw new Error(
- "PINECONE_ENVIRONMENT and PINECONE_API_KEY and PINECONE_INDEX must be set"
- );
+if (!process.env.PINECONE_API_KEY || !process.env.PINECONE_INDEX) {
+ throw new Error("PINECONE_API_KEY and PINECONE_INDEX must be set");
}
const pinecone = new Pinecone();
diff --git a/libs/langchain-community/package.json b/libs/langchain-community/package.json
index 61ee6c6410f9..8de0a386cb34 100644
--- a/libs/langchain-community/package.json
+++ b/libs/langchain-community/package.json
@@ -91,7 +91,6 @@
"@neondatabase/serverless": "^0.9.1",
"@notionhq/client": "^2.2.10",
"@opensearch-project/opensearch": "^2.2.0",
- "@pinecone-database/pinecone": "^1.1.0",
"@planetscale/database": "^1.8.0",
"@premai/prem-sdk": "^0.3.25",
"@qdrant/js-client-rest": "^1.8.2",
diff --git a/libs/langchain-pinecone/jest.config.cjs b/libs/langchain-pinecone/jest.config.cjs
index a49a8832a349..daf6c5865cbe 100644
--- a/libs/langchain-pinecone/jest.config.cjs
+++ b/libs/langchain-pinecone/jest.config.cjs
@@ -17,5 +17,5 @@ module.exports = {
setupFiles: ["dotenv/config"],
testTimeout: 20_000,
passWithNoTests: true,
- collectCoverageFrom: ["src/**/*.ts"]
-};
+ collectCoverageFrom: ["src/**/*.ts"],
+ };
diff --git a/libs/langchain-pinecone/package.json b/libs/langchain-pinecone/package.json
index 2f6e4cef205f..106855d9ffd2 100644
--- a/libs/langchain-pinecone/package.json
+++ b/libs/langchain-pinecone/package.json
@@ -32,7 +32,7 @@
"author": "Pinecone, Inc",
"license": "MIT",
"dependencies": {
- "@pinecone-database/pinecone": "^3.0.0 || ^4.0.0",
+ "@pinecone-database/pinecone": "^4.0.0",
"flat": "^5.0.2",
"uuid": "^10.0.0"
},
diff --git a/libs/langchain-pinecone/src/client.ts b/libs/langchain-pinecone/src/client.ts
new file mode 100644
index 000000000000..f1eabb825db5
--- /dev/null
+++ b/libs/langchain-pinecone/src/client.ts
@@ -0,0 +1,16 @@
+import { Pinecone, PineconeConfiguration } from "@pinecone-database/pinecone";
+import { getEnvironmentVariable } from "@langchain/core/utils/env";
+
+export function getPineconeClient(config?: PineconeConfiguration): Pinecone {
+ if (
+ getEnvironmentVariable("PINECONE_API_KEY") === undefined ||
+ getEnvironmentVariable("PINECONE_API_KEY") === ""
+ ) {
+ throw new Error("PINECONE_API_KEY must be set in environment");
+ }
+ if (!config) {
+ return new Pinecone();
+ } else {
+ return new Pinecone(config);
+ }
+}
diff --git a/libs/langchain-pinecone/src/embeddings.ts b/libs/langchain-pinecone/src/embeddings.ts
new file mode 100644
index 000000000000..6d0d04226ed8
--- /dev/null
+++ b/libs/langchain-pinecone/src/embeddings.ts
@@ -0,0 +1,139 @@
+/* eslint-disable arrow-body-style */
+
+import { Embeddings, type EmbeddingsParams } from "@langchain/core/embeddings";
+import {
+ EmbeddingsList,
+ Pinecone,
+ PineconeConfiguration,
+} from "@pinecone-database/pinecone";
+import { getPineconeClient } from "./client.js";
+
+/* PineconeEmbeddingsParams holds the optional fields a user can pass to a Pinecone embedding model.
+ * @param model - Model to use to generate embeddings. Default is "multilingual-e5-large".
+ * @param params - Additional parameters to pass to the embedding model. Note: parameters are model-specific. Read
+ * more about model-specific parameters in the [Pinecone
+ * documentation](https://docs.pinecone.io/guides/inference/understanding-inference#model-specific-parameters).
+ * */
+export interface PineconeEmbeddingsParams extends EmbeddingsParams {
+ model?: string; // Model to use to generate embeddings
+ params?: Record; // Additional parameters to pass to the embedding model
+}
+
+/* PineconeEmbeddings generates embeddings using the Pinecone Inference API. */
+export class PineconeEmbeddings
+ extends Embeddings
+ implements PineconeEmbeddingsParams
+{
+ client: Pinecone;
+
+ model: string;
+
+ params: Record;
+
+ constructor(
+ fields?: Partial & Partial
+ ) {
+ const defaultFields = { maxRetries: 3, ...fields };
+ super(defaultFields);
+
+ if (defaultFields.apiKey) {
+ const config = {
+ apiKey: defaultFields.apiKey,
+ controllerHostUrl: defaultFields.controllerHostUrl,
+ fetchApi: defaultFields.fetchApi,
+ additionalHeaders: defaultFields.additionalHeaders,
+ sourceTag: defaultFields.sourceTag,
+ } as PineconeConfiguration;
+ this.client = getPineconeClient(config);
+ } else {
+ this.client = getPineconeClient();
+ }
+
+ if (!defaultFields.model) {
+ this.model = "multilingual-e5-large";
+ } else {
+ this.model = defaultFields.model;
+ }
+
+ const defaultParams = { inputType: "passage" };
+
+ if (defaultFields.params) {
+ this.params = { ...defaultFields.params, ...defaultParams };
+ } else {
+ this.params = defaultParams;
+ }
+ }
+
+ /* Generate embeddings for a list of input strings using a specified embedding model.
+ *
+ * @param texts - List of input strings for which to generate embeddings.
+ * */
+ async embedDocuments(texts: string[]): Promise {
+ if (texts.length === 0) {
+ throw new Error(
+ "At least one document is required to generate embeddings"
+ );
+ }
+
+ let embeddings;
+ if (this.params) {
+ embeddings = await this.caller.call(async () => {
+ const result: EmbeddingsList = await this.client.inference.embed(
+ this.model,
+ texts,
+ this.params
+ );
+ return result;
+ });
+ } else {
+ embeddings = await this.caller.call(async () => {
+ const result: EmbeddingsList = await this.client.inference.embed(
+ this.model,
+ texts,
+ {}
+ );
+ return result;
+ });
+ }
+
+ const embeddingsList: number[][] = [];
+
+ for (let i = 0; i < embeddings.length; i += 1) {
+ if (embeddings[i].values) {
+ embeddingsList.push(embeddings[i].values as number[]);
+ }
+ }
+ return embeddingsList;
+ }
+
+ /* Generate embeddings for a given query string using a specified embedding model.
+ * @param text - Query string for which to generate embeddings.
+ * */
+ async embedQuery(text: string): Promise {
+ // Change inputType to query-specific param for multilingual-e5-large embedding model
+ this.params.inputType = "query";
+
+ if (!text) {
+ throw new Error("No query passed for which to generate embeddings");
+ }
+ let embeddings: EmbeddingsList;
+ if (this.params) {
+ embeddings = await this.caller.call(async () => {
+ return await this.client.inference.embed(
+ this.model,
+ [text],
+ this.params
+ );
+ });
+ } else {
+ embeddings = await this.caller.call(async () => {
+ return await this.client.inference.embed(this.model, [text], {});
+ });
+ }
+ if (embeddings[0].values) {
+ return embeddings[0].values as number[];
+ } else {
+ return [];
+ }
+ }
+}
diff --git a/libs/langchain-pinecone/src/index.ts b/libs/langchain-pinecone/src/index.ts
index d464092ede5e..f3a0f54563c6 100644
--- a/libs/langchain-pinecone/src/index.ts
+++ b/libs/langchain-pinecone/src/index.ts
@@ -1,2 +1,3 @@
export * from "./vectorstores.js";
export * from "./translator.js";
+export * from "./embeddings.js";
diff --git a/libs/langchain-pinecone/src/tests/client.int.test.ts b/libs/langchain-pinecone/src/tests/client.int.test.ts
new file mode 100644
index 000000000000..b52beb2cf2b8
--- /dev/null
+++ b/libs/langchain-pinecone/src/tests/client.int.test.ts
@@ -0,0 +1,39 @@
+import { Pinecone } from "@pinecone-database/pinecone";
+import { getPineconeClient } from "../client.js";
+
+describe("Tests for getPineconeClient", () => {
+ test("Happy path for getPineconeClient with and without `config` obj passed", async () => {
+ const client = getPineconeClient();
+ expect(client).toBeInstanceOf(Pinecone);
+ expect(client).toHaveProperty("config"); // Config is always set to *at least* the user's api key
+
+ const clientWithConfig = getPineconeClient({
+ // eslint-disable-next-line no-process-env
+ apiKey: process.env.PINECONE_API_KEY!,
+ additionalHeaders: { header: "value" },
+ });
+ expect(clientWithConfig).toBeInstanceOf(Pinecone);
+ expect(client).toHaveProperty("config"); // Unfortunately cannot assert on contents of config b/c it's a private
+ // attribute of the Pinecone class
+ });
+
+ test("Unhappy path: expect getPineconeClient to throw error if reset PINECONE_API_KEY to empty string", async () => {
+ // eslint-disable-next-line no-process-env
+ const originalApiKey = process.env.PINECONE_API_KEY;
+ try {
+ // eslint-disable-next-line no-process-env
+ process.env.PINECONE_API_KEY = "";
+ const errorThrown = async () => {
+ getPineconeClient();
+ };
+ await expect(errorThrown).rejects.toThrow(Error);
+ await expect(errorThrown).rejects.toThrow(
+ "PINECONE_API_KEY must be set in environment"
+ );
+ } finally {
+ // Restore the original value of PINECONE_API_KEY
+ // eslint-disable-next-line no-process-env
+ process.env.PINECONE_API_KEY = originalApiKey;
+ }
+ });
+});
diff --git a/libs/langchain-pinecone/src/tests/client.test.ts b/libs/langchain-pinecone/src/tests/client.test.ts
new file mode 100644
index 000000000000..60b7f6dcb348
--- /dev/null
+++ b/libs/langchain-pinecone/src/tests/client.test.ts
@@ -0,0 +1,15 @@
+import { getPineconeClient } from "../client.js";
+
+describe("Tests for getPineconeClient", () => {
+ test("Confirm getPineconeClient throws error when PINECONE_API_KEY is not set", async () => {
+ /* eslint-disable-next-line no-process-env */
+ process.env.PINECONE_API_KEY = "";
+ const errorThrown = async () => {
+ getPineconeClient();
+ };
+ await expect(errorThrown).rejects.toThrow(Error);
+ await expect(errorThrown).rejects.toThrow(
+ "PINECONE_API_KEY must be set in environment"
+ );
+ });
+});
diff --git a/libs/langchain-pinecone/src/tests/embeddings.int.test.ts b/libs/langchain-pinecone/src/tests/embeddings.int.test.ts
new file mode 100644
index 000000000000..980e32775bca
--- /dev/null
+++ b/libs/langchain-pinecone/src/tests/embeddings.int.test.ts
@@ -0,0 +1,59 @@
+import { PineconeEmbeddings } from "../embeddings.js";
+
+describe("Integration tests for Pinecone embeddings", () => {
+ test("Happy path: defaults for both embedDocuments and embedQuery", async () => {
+ const model = new PineconeEmbeddings();
+ expect(model.model).toBe("multilingual-e5-large");
+ expect(model.params).toEqual({ inputType: "passage" });
+
+ const docs = ["hello", "world"];
+ const embeddings = await model.embedDocuments(docs);
+ expect(embeddings.length).toBe(docs.length);
+
+ const query = "hello";
+ const queryEmbedding = await model.embedQuery(query);
+ expect(queryEmbedding.length).toBeGreaterThan(0);
+ });
+
+ test("Happy path: custom `params` obj passed to embedDocuments and embedQuery", async () => {
+ const model = new PineconeEmbeddings({
+ params: { customParam: "value" },
+ });
+ expect(model.model).toBe("multilingual-e5-large");
+ expect(model.params).toEqual({
+ inputType: "passage",
+ customParam: "value",
+ });
+
+ const docs = ["hello", "world"];
+ const embeddings = await model.embedDocuments(docs);
+ expect(embeddings.length).toBe(docs.length);
+ expect(embeddings[0].length).toBe(1024); // Assert correct dims on random doc
+ expect(model.model).toBe("multilingual-e5-large");
+ expect(model.params).toEqual({
+ inputType: "passage", // Maintain default inputType for docs
+ customParam: "value",
+ });
+
+ const query = "hello";
+ const queryEmbedding = await model.embedQuery(query);
+ expect(model.model).toBe("multilingual-e5-large");
+ expect(queryEmbedding.length).toBe(1024);
+ expect(model.params).toEqual({
+ inputType: "query", // Change inputType for query
+ customParam: "value",
+ });
+ });
+
+ test("Unhappy path: embedDocuments and embedQuery throw when empty objs are passed", async () => {
+ const model = new PineconeEmbeddings();
+ await expect(model.embedDocuments([])).rejects.toThrow();
+ await expect(model.embedQuery("")).rejects.toThrow();
+ });
+
+ test("Unhappy path: PineconeEmbeddings throws when invalid model is passed", async () => {
+ const model = new PineconeEmbeddings({ model: "invalid-model" });
+ await expect(model.embedDocuments([])).rejects.toThrow();
+ await expect(model.embedQuery("")).rejects.toThrow();
+ });
+});
diff --git a/libs/langchain-pinecone/src/tests/embeddings.test.ts b/libs/langchain-pinecone/src/tests/embeddings.test.ts
new file mode 100644
index 000000000000..5b203e758a5b
--- /dev/null
+++ b/libs/langchain-pinecone/src/tests/embeddings.test.ts
@@ -0,0 +1,48 @@
+import { PineconeEmbeddings } from "../embeddings.js";
+
+beforeAll(() => {
+ // eslint-disable-next-line no-process-env
+ process.env.PINECONE_API_KEY = "test-api-key";
+});
+
+describe("Tests for the PineconeEmbeddings class", () => {
+ test("Confirm embedDocuments method throws error when an empty array is passed", async () => {
+ const model = new PineconeEmbeddings();
+ const errorThrown = async () => {
+ await model.embedDocuments([]);
+ };
+ await expect(errorThrown).rejects.toThrow(Error);
+ await expect(errorThrown).rejects.toThrowError(
+ "At least one document is required to generate embeddings"
+ );
+ });
+
+ test("Confirm embedQuery method throws error when an empty string is passed", async () => {
+ const model = new PineconeEmbeddings();
+ const errorThrown = async () => {
+ await model.embedQuery("");
+ };
+ await expect(errorThrown).rejects.toThrow(Error);
+ await expect(errorThrown).rejects.toThrowError(
+ "No query passed for which to generate embeddings"
+ );
+ });
+
+ test("Confirm instance defaults are set when no args are passed", async () => {
+ const model = new PineconeEmbeddings();
+ expect(model.model).toBe("multilingual-e5-large");
+ expect(model.params).toEqual({ inputType: "passage" });
+ });
+
+ test("Confirm instance sets custom model and params when provided", () => {
+ const customModel = new PineconeEmbeddings({
+ model: "custom-model",
+ params: { customParam: "value" },
+ });
+ expect(customModel.model).toBe("custom-model");
+ expect(customModel.params).toEqual({
+ inputType: "passage",
+ customParam: "value",
+ });
+ });
+});
diff --git a/libs/langchain-pinecone/src/tests/translator.int.test.ts b/libs/langchain-pinecone/src/tests/translator.int.test.ts
index 69a2b8fca663..efae21d505e0 100644
--- a/libs/langchain-pinecone/src/tests/translator.int.test.ts
+++ b/libs/langchain-pinecone/src/tests/translator.int.test.ts
@@ -109,14 +109,8 @@ describe("Pinecone self query", () => {
},
];
- if (
- !process.env.PINECONE_API_KEY ||
- !process.env.PINECONE_ENVIRONMENT ||
- !testIndexName
- ) {
- throw new Error(
- "PINECONE_ENVIRONMENT and PINECONE_API_KEY and PINECONE_INDEX must be set"
- );
+ if (!process.env.PINECONE_API_KEY || !testIndexName) {
+ throw new Error("PINECONE_API_KEY and PINECONE_INDEX must be set");
}
const embeddings = new OpenAIEmbeddings();
@@ -268,14 +262,8 @@ describe("Pinecone self query", () => {
},
];
- if (
- !process.env.PINECONE_API_KEY ||
- !process.env.PINECONE_ENVIRONMENT ||
- !testIndexName
- ) {
- throw new Error(
- "PINECONE_ENVIRONMENT and PINECONE_API_KEY and PINECONE_INDEX must be set"
- );
+ if (!process.env.PINECONE_API_KEY || !testIndexName) {
+ throw new Error("PINECONE_API_KEY and PINECONE_INDEX must be set");
}
const embeddings = new OpenAIEmbeddings();
diff --git a/libs/langchain-pinecone/src/vectorstores.ts b/libs/langchain-pinecone/src/vectorstores.ts
index 3e4db8bbc30c..8601c1570e8f 100644
--- a/libs/langchain-pinecone/src/vectorstores.ts
+++ b/libs/langchain-pinecone/src/vectorstores.ts
@@ -88,7 +88,7 @@ export type PineconeDeleteParams = {
*
* const pinecone = new PineconeClient();
*
- * // Will automatically read the PINECONE_API_KEY and PINECONE_ENVIRONMENT env vars
+ * // Will automatically read the PINECONE_API_KEY env var
* const pineconeIndex = pinecone.Index(process.env.PINECONE_INDEX!);
*
* const embeddings = new OpenAIEmbeddings({
diff --git a/yarn.lock b/yarn.lock
index 42dd445ba3fe..ee400067548a 100644
--- a/yarn.lock
+++ b/yarn.lock
@@ -9377,22 +9377,6 @@ __metadata:
languageName: node
linkType: hard
-"@edge-runtime/primitives@npm:4.0.2":
- version: 4.0.2
- resolution: "@edge-runtime/primitives@npm:4.0.2"
- checksum: 66ccdadb8bcb02bf7d75d75bc6a611c2d515b09818b8cc4f524944aba58b735aff8a76ba52981a04f89fe616f801d1d947a9b1637a201a3b32f6eca74a6f7001
- languageName: node
- linkType: hard
-
-"@edge-runtime/types@npm:^2.2.3":
- version: 2.2.4
- resolution: "@edge-runtime/types@npm:2.2.4"
- dependencies:
- "@edge-runtime/primitives": 4.0.2
- checksum: 51b5b1cf8f462dd77038357e1b72737e0e8f37c6e7aa47e0b9cd3aebc81004ad5e41c697d14a66171e5acfddf8bf5a5fc6e76c01dde60daccb5703310a4c4446
- languageName: node
- linkType: hard
-
"@elastic/elasticsearch@npm:^8.4.0":
version: 8.8.1
resolution: "@elastic/elasticsearch@npm:8.8.1"
@@ -11530,7 +11514,6 @@ __metadata:
"@neondatabase/serverless": ^0.9.1
"@notionhq/client": ^2.2.10
"@opensearch-project/opensearch": ^2.2.0
- "@pinecone-database/pinecone": ^1.1.0
"@planetscale/database": ^1.8.0
"@premai/prem-sdk": ^0.3.25
"@qdrant/js-client-rest": ^1.8.2
@@ -12608,7 +12591,7 @@ __metadata:
"@langchain/core": "workspace:*"
"@langchain/openai": "workspace:*"
"@langchain/scripts": ">=0.1.0 <0.2.0"
- "@pinecone-database/pinecone": ^3.0.0 || ^4.0.0
+ "@pinecone-database/pinecone": ^4.0.0
"@swc/core": ^1.3.90
"@swc/jest": ^0.2.29
"@tsconfig/recommended": ^1.0.3
@@ -14130,22 +14113,7 @@ __metadata:
languageName: node
linkType: hard
-"@pinecone-database/pinecone@npm:^1.1.0":
- version: 1.1.0
- resolution: "@pinecone-database/pinecone@npm:1.1.0"
- dependencies:
- "@edge-runtime/types": ^2.2.3
- "@sinclair/typebox": ^0.29.0
- "@types/node": ^18.11.17
- ajv: ^8.12.0
- cross-fetch: ^3.1.5
- encoding: ^0.1.13
- typescript: ^4.9.4
- checksum: 5ff066c23006c6ad8eb3ae5add5aa2633adacf48c539a498f6d97798fb9640254a2ede99f7e47aaff3d885e3df5168ed193990d846acd9bebded7140f18c8eb1
- languageName: node
- linkType: hard
-
-"@pinecone-database/pinecone@npm:^3.0.0 || ^4.0.0, @pinecone-database/pinecone@npm:^4.0.0":
+"@pinecone-database/pinecone@npm:^4.0.0":
version: 4.0.0
resolution: "@pinecone-database/pinecone@npm:4.0.0"
dependencies:
@@ -14822,13 +14790,6 @@ __metadata:
languageName: node
linkType: hard
-"@sinclair/typebox@npm:^0.29.0":
- version: 0.29.6
- resolution: "@sinclair/typebox@npm:0.29.6"
- checksum: 02c2ee9e8bcb4e03a2497fc208a5a2ae4f4991347e6f2b2a217027b33eb8ca2bc74e80bfb2654da537453e9e5c25d59dbad2d32d4fe47ffec357330bf343bf7a
- languageName: node
- linkType: hard
-
"@sindresorhus/is@npm:^0.14.0":
version: 0.14.0
resolution: "@sindresorhus/is@npm:0.14.0"
@@ -19363,13 +19324,6 @@ __metadata:
languageName: node
linkType: hard
-"@types/node@npm:^18.11.17":
- version: 18.18.4
- resolution: "@types/node@npm:18.18.4"
- checksum: 4901e91c4cc479bb58acbcd79236a97a0ad6db4a53cb1f4ba4cf32af15324c61b16faa6e31c1b09bf538a20feb5f5274239157ce5237f5741db0b9ab71e69c52
- languageName: node
- linkType: hard
-
"@types/node@npm:^18.11.18":
version: 18.16.19
resolution: "@types/node@npm:18.16.19"
@@ -20958,7 +20912,7 @@ __metadata:
languageName: node
linkType: hard
-"ajv@npm:^8.0.0, ajv@npm:^8.12.0, ajv@npm:^8.6.3, ajv@npm:^8.9.0":
+"ajv@npm:^8.0.0, ajv@npm:^8.6.3, ajv@npm:^8.9.0":
version: 8.12.0
resolution: "ajv@npm:8.12.0"
dependencies: