Skip to main content

TogetherAIEmbeddings integration

还没翻译译自英文（docs.langchain.com）本页 27 段都还没译

This will help you get started with TogetherAIEmbeddings embedding models using LangChain. For detailed documentation on TogetherAIEmbeddings features and configuration options, please refer to the API reference.

Overview

Integration details

ClassPackageLocalPy supportDownloadsVersion
TogetherAIEmbeddings@langchain/together-ai❌✅NPM - DownloadsNPM - Version

Setup

To access Together AI embedding models, create a Together account, get an API key, and install the @langchain/together-ai integration package.

Credentials

You can sign up for a Together account and create an API key. Set the TOGETHER_AI_API_KEY environment variable:

export TOGETHER_AI_API_KEY="your-api-key"

If you want to get automated tracing of your model calls you can also set your LangSmith API key by uncommenting below:

# export LANGSMITH_TRACING="true"
# export LANGSMITH_API_KEY="your-api-key"

Installation

The LangChain TogetherAIEmbeddings integration lives in the @langchain/together-ai package:

npm install @langchain/together-ai @langchain/core
yarn add @langchain/together-ai @langchain/core
pnpm add @langchain/together-ai @langchain/core

Instantiation

Instantiate the embeddings model:

import { TogetherAIEmbeddings } from "@langchain/together-ai";

const embeddings = new TogetherAIEmbeddings({
  model: "togethercomputer/m2-bert-80M-8k-retrieval", // Default value
});

Indexing and retrieval

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 Learn tab.

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.

// Create a vector store with a sample text
import { MemoryVectorStore } from "@langchain/classic/vectorstores/memory";

const text = "LangChain is the framework for building context-aware reasoning applications";

const vectorstore = await MemoryVectorStore.fromDocuments(
  [{ pageContent: text, metadata: {} }],
  embeddings,
);

// Use the vector store as a retriever that returns a single document
const retriever = vectorstore.asRetriever(1);

// Retrieve the most similar text
const retrievedDocuments = await retriever.invoke("What is LangChain?");

retrievedDocuments[0].pageContent;
LangChain is the framework for building context-aware reasoning applications

Direct usage

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.

You can directly call these methods to get embeddings for your own use cases.

Embed single texts

You can embed queries for search with embedQuery. This generates a vector representation specific to the query:

const singleVector = await embeddings.embedQuery(text);

console.log(singleVector.slice(0, 100));
[
     0.3812227, -0.052848946,  -0.10564975,   0.03480297,    0.2878488,
  0.0084609175,   0.11605915,   0.05303011,   0.14711718,  -0.14407106,
   -0.29865336,  -0.15807179, -0.068397366,   -0.2708063,  0.056596708,
   -0.07656515,  0.052995138,  -0.11275427,  0.028096694,  0.123501234,
  -0.039519835,   0.12148692,  -0.12820457,   0.15691335,  0.033519063,
   -0.27026987,  -0.08460162,  -0.23792154,    -0.234982,  -0.05786798,
   0.016467346,  -0.17168592, -0.060787182,  0.038752213,  -0.08169927,
    0.09327062,   0.29490772,    0.0167866,  -0.32224452,   -0.2037822,
   -0.10284172, -0.124050565,   0.25344968,  -0.06275548,  -0.14180769,
  0.0046709594,  0.073105976,   0.12004031,   0.19224276, -0.022589967,
   0.102790825,    0.1138286, -0.057701062, -0.050010648,   -0.1632584,
   -0.18942119,  -0.12018798,   0.15288158,   0.07941474,   0.10440051,
   -0.13257962,  -0.19282033,  0.044656333,   0.13560675, -0.068929024,
   0.028590716,  0.055663664,   0.04652713,  0.014936657,  0.120679885,
   0.053866718,  -0.16296014,  0.119450666,  -0.29559663,  0.008097747,
    0.07380408,  -0.09010084,   -0.0687739,  -0.08575685,  -0.07202606,
    0.18868081,  -0.08392917,  0.014016109,   0.15435852, -0.030115498,
   -0.16927013,   0.02836557, -0.050763763,    0.0840437,  -0.22718845,
   0.111397505,  0.033395614, -0.123287566,   -0.2111604,   -0.1580479,
    0.05520573,   -0.1422921,   0.08828953,  0.051058788,  -0.13312188
]

Embed multiple texts

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:

const text2 = "LangGraph is a library for building stateful, multi-actor applications with LLMs";

const vectors = await embeddings.embedDocuments([text, text2]);

console.log(vectors[0].slice(0, 100));
console.log(vectors[1].slice(0, 100));
[
     0.3812227, -0.052848946,  -0.10564975,   0.03480297,    0.2878488,
  0.0084609175,   0.11605915,   0.05303011,   0.14711718,  -0.14407106,
   -0.29865336,  -0.15807179, -0.068397366,   -0.2708063,  0.056596708,
   -0.07656515,  0.052995138,  -0.11275427,  0.028096694,  0.123501234,
  -0.039519835,   0.12148692,  -0.12820457,   0.15691335,  0.033519063,
   -0.27026987,  -0.08460162,  -0.23792154,    -0.234982,  -0.05786798,
   0.016467346,  -0.17168592, -0.060787182,  0.038752213,  -0.08169927,
    0.09327062,   0.29490772,    0.0167866,  -0.32224452,   -0.2037822,
   -0.10284172, -0.124050565,   0.25344968,  -0.06275548,  -0.14180769,
  0.0046709594,  0.073105976,   0.12004031,   0.19224276, -0.022589967,
   0.102790825,    0.1138286, -0.057701062, -0.050010648,   -0.1632584,
   -0.18942119,  -0.12018798,   0.15288158,   0.07941474,   0.10440051,
   -0.13257962,  -0.19282033,  0.044656333,   0.13560675, -0.068929024,
   0.028590716,  0.055663664,   0.04652713,  0.014936657,  0.120679885,
   0.053866718,  -0.16296014,  0.119450666,  -0.29559663,  0.008097747,
    0.07380408,  -0.09010084,   -0.0687739,  -0.08575685,  -0.07202606,
    0.18868081,  -0.08392917,  0.014016109,   0.15435852, -0.030115498,
   -0.16927013,   0.02836557, -0.050763763,    0.0840437,  -0.22718845,
   0.111397505,  0.033395614, -0.123287566,   -0.2111604,   -0.1580479,
    0.05520573,   -0.1422921,   0.08828953,  0.051058788,  -0.13312188
]
[
   0.066308185, -0.032866564,  0.115751594,   0.19082588,      0.14017,
   -0.26976448, -0.056340694,  -0.26923394,    0.2548541,  -0.27271318,
    -0.2244126,   0.07949589,  -0.27710953,  -0.17993368,   0.09681616,
   -0.08692256,   0.22127126,  -0.14512022,  -0.18016525,   0.14892976,
    -0.0526347, -0.008140617,   -0.2916987,   0.23706906,  -0.38488507,
   -0.35881752,   0.09276949,  -0.07051063,  -0.07778231,   0.12552947,
    0.06256748,  -0.25832427,  0.025054429,   -0.1451448,   -0.2662871,
    0.13676351,  -0.07413256,   0.14966589,  -0.39968985,   0.15542287,
   -0.13107607,   0.02761394,  0.108077586,  -0.12076956,     0.128296,
   -0.05625126,   0.15723586, -0.056932643,   0.23720805,   0.23993455,
  -0.035553705, -0.053907514,  -0.11852807,   0.07005695,  -0.06317475,
   0.070009425,     0.284697,    0.2212059,  0.018890115,   0.16924675,
    0.21651487,   0.07259682,    0.1328156,    0.3261852,    0.1914124,
   -0.10120423,   0.03450111,  -0.22588971,  -0.04458192,   0.24116798,
  -0.021830376,  -0.30731413,   0.08586451, -0.058835756, 0.0010347435,
  0.0031927782,  -0.09403646,  -0.22608931,   0.15865424,   0.15738021,
    0.23582733,    0.1714161,    0.1585189,  -0.18085755,  0.019376995,
  -0.026587496, -0.017079154,  -0.04588549, -0.047336094, -0.082413346,
    -0.1114185,  -0.05403556,   0.12438637,  -0.20476522,     0.073182,
   -0.12210378, -0.010543863,  -0.09767598,    0.1057683, -0.050204434
]

API reference

For detailed documentation of all TogetherAIEmbeddings features and configurations head to the API reference.

AI 翻译查看英文原文