langchain.js
    Preparing search index...

    Function initChatModel

    Initialize a ChatModel from the model name and provider. Must have the integration package corresponding to the model provider installed.

    The input type for the model.

    Call options for the model.

    The name of the model, e.g. "gpt-4", "claude-3-opus-20240229". Can be prefixed with the model provider, e.g. "openai:gpt-4", "anthropic:claude-3-opus-20240229".

    Additional configuration options.

    The model provider. Supported values include:

    • openai (@langchain/openai)
    • anthropic (@langchain/anthropic)
    • azure_openai (@langchain/openai)
    • google-vertexai (@langchain/google-vertexai)
    • google-vertexai-web (@langchain/google-vertexai-web)
    • google-genai (@langchain/google-genai)
    • bedrock (@langchain/aws)
    • cohere (@langchain/cohere)
    • fireworks (@langchain/community/chat_models/fireworks)
    • together (@langchain/community/chat_models/togetherai)
    • mistralai (@langchain/mistralai)
    • groq (@langchain/groq)
    • ollama (@langchain/ollama)
    • perplexity (@langchain/community/chat_models/perplexity)
    • cerebras (@langchain/cerebras)
    • deepseek (@langchain/deepseek)
    • xai (@langchain/xai)

    Which model parameters are configurable:

    • undefined: No configurable fields.
    • "any": All fields are configurable. (See Security Note in description)
    • string[]: Specified fields are configurable.

    Prefix for configurable fields at runtime.

    Overrides the profiling information for the model. If not provided, the profile will be inferred from the inner model instance.

    Additional keyword args to pass to the ChatModel constructor.

    A class which extends BaseChatModel.

    If modelProvider cannot be inferred or isn't supported.

    If the model provider integration package is not installed.

    import { initChatModel } from "langchain/chat_models/universal";

    const gpt4 = await initChatModel("openai:gpt-4", {
      temperature: 0.25,
    });
    const gpt4Result = await gpt4.invoke("what's your name");

    const claude = await initChatModel("anthropic:claude-3-opus-20240229", {
      temperature: 0.25,
    });
    const claudeResult = await claude.invoke("what's your name");

    const gemini = await initChatModel("gemini-1.5-pro", {
      modelProvider: "google-vertexai",
      temperature: 0.25,
    });
    const geminiResult = await gemini.invoke("what's your name");

    import { initChatModel } from "langchain/chat_models/universal";

    const configurableModel = await initChatModel(undefined, {
      temperature: 0,
      configurableFields: ["model", "apiKey"],
    });

    const gpt4Result = await configurableModel.invoke("what's your name", {
      configurable: {
        model: "gpt-4",
      },
    });

    const claudeResult = await configurableModel.invoke("what's your name", {
      configurable: {
        model: "claude-3-5-sonnet-20240620",
      },
    });

    import { initChatModel } from "langchain/chat_models/universal";

    const configurableModelWithDefault = await initChatModel("gpt-4", {
      modelProvider: "openai",
      configurableFields: "any",
      configPrefix: "foo",
      temperature: 0,
    });

    const openaiResult = await configurableModelWithDefault.invoke(
      "what's your name",
      {
        configurable: {
          foo_apiKey: process.env.OPENAI_API_KEY,
        },
      }
    );

    const claudeResult = await configurableModelWithDefault.invoke(
      "what's your name",
      {
        configurable: {
          foo_model: "claude-3-5-sonnet-20240620",
          foo_modelProvider: "anthropic",
          foo_temperature: 0.6,
          foo_apiKey: process.env.ANTHROPIC_API_KEY,
        },
      }
    );

    import { initChatModel } from "langchain/chat_models/universal";
    import { z } from "zod/v3";
    import { tool } from "@langchain/core/tools";

    const getWeatherTool = tool(
      (input) => {
        // Do something with the input
        return JSON.stringify(input);
      },
      {
        schema: z
          .object({
            location: z
              .string()
              .describe("The city and state, e.g. San Francisco, CA"),
          })
          .describe("Get the current weather in a given location"),
        name: "GetWeather",
        description: "Get the current weather in a given location",
      }
    );

    const getPopulationTool = tool(
      (input) => {
        // Do something with the input
        return JSON.stringify(input);
      },
      {
        schema: z
          .object({
            location: z
              .string()
              .describe("The city and state, e.g. San Francisco, CA"),
          })
          .describe("Get the current population in a given location"),
        name: "GetPopulation",
        description: "Get the current population in a given location",
      }
    );

    const configurableModel = await initChatModel("gpt-4", {
      configurableFields: ["model", "modelProvider", "apiKey"],
      temperature: 0,
    });

    const configurableModelWithTools = configurableModel.bindTools([
      getWeatherTool,
      getPopulationTool,
    ]);

    const configurableToolResult = await configurableModelWithTools.invoke(
      "Which city is hotter today and which is bigger: LA or NY?",
      {
        configurable: {
          apiKey: process.env.OPENAI_API_KEY,
        },
      }
    );

    const configurableToolResult2 = await configurableModelWithTools.invoke(
      "Which city is hotter today and which is bigger: LA or NY?",
      {
        configurable: {
          model: "claude-3-5-sonnet-20240620",
          apiKey: process.env.ANTHROPIC_API_KEY,
        },
      }
    );

    import { initChatModel } from "langchain/chat_models/universal";

    const model = await initChatModel("gpt-4o-mini", {
      profile: {
        maxInputTokens: 100000,
      },
    });

    @description
    This function initializes a ChatModel based on the provided model name and provider.
    It supports various model providers and allows for runtime configuration of model parameters.

    Security Note: Setting `configurableFields` to "any" means fields like apiKey, baseUrl, etc.
    can be altered at runtime, potentially redirecting model requests to a different service/user.
    Make sure that if you're accepting untrusted configurations, you enumerate the
    `configurableFields` explicitly.

    The function will attempt to infer the model provider from the model name if not specified.
    Certain model name prefixes are associated with specific providers:
    - gpt-3... or gpt-4... -> openai
    - claude... -> anthropic
    - amazon.... -> bedrock
    - gemini... -> google-vertexai
    - command... -> cohere
    - accounts/fireworks... -> fireworks

    @since 0.2.11
    @version

    Settings

    Member Visibility