Class ChatCohere<CallOptions>

Integration for Cohere chat models.

Setup: Install @langchain/cohere and set a environment variable called COHERE_API_KEY.

npm install @langchain/cohere
export COHERE_API_KEY="your-api-key"

Constructor args

Runtime args

Runtime args can be passed as the second argument to any of the base runnable methods .invoke. .stream, .batch, etc. They can also be passed via .bind, or the second arg in .bindTools, like shown in the examples below:

// When calling `.bind`, call options should be passed via the first argument
const llmWithArgsBound = llm.bind({
  stop: ["\n"],
  tools: [...],
});

// When calling `.bindTools`, call options should be passed via the second argument
const llmWithTools = llm.bindTools(
  [...],
  {
    stop: ["\n"],
  }
);

Examples

Instantiate 
import { ChatCohere } from '@langchain/cohere';

const llm = new ChatCohere({
  model: "command-r-plus",
  temperature: 0,
  // other params...
});

Invoking 
const input = `Translate "I love programming" into French.`;

// Models also accept a list of chat messages or a formatted prompt
const result = await llm.invoke(input);
console.log(result);

AIMessage {
  "content": "\"J'adore programmer.\"",
  "additional_kwargs": {
    ...
  },
  "response_metadata": {
    "estimatedTokenUsage": {
      "completionTokens": 6,
      "promptTokens": 75,
      "totalTokens": 81
    },
    "response_id": "54cebd43-737f-458b-bff4-01b220eaf373",
    "generationId": "48a567da-0f88-4606-bba6-becbeee464bd",
    "chatHistory": [
      {
        "role": "USER",
        "message": "Translate \"I love programming\" into French."
      },
      {
        "role": "CHATBOT",
        "message": "\"J'adore programmer.\""
      }
    ],
    "finishReason": "COMPLETE",
    "meta": {
      "apiVersion": {
        "version": "1"
      },
      "billedUnits": {
        "inputTokens": 9,
        "outputTokens": 6
      },
      "tokens": {
        "inputTokens": 75,
        "outputTokens": 6
      }
    }
  },
  "tool_calls": [],
  "invalid_tool_calls": [],
  "usage_metadata": {
    "input_tokens": 75,
    "output_tokens": 6,
    "total_tokens": 81
  }
}

Streaming Chunks 
for await (const chunk of await llm.stream(input)) {
  console.log(chunk);
}

AIMessageChunk {
  "content": "",
  "additional_kwargs": {
    "eventType": "stream-start",
    "is_finished": false,
    "generationId": "d62c8989-8af5-4357-af79-4ea8e6eb2baa"
  },
  "response_metadata": {
    "eventType": "stream-start",
    "is_finished": false,
    "generationId": "d62c8989-8af5-4357-af79-4ea8e6eb2baa"
  },
  "tool_calls": [],
  "tool_call_chunks": [],
  "invalid_tool_calls": []
}
AIMessageChunk {
  "content": "\"",
  "additional_kwargs": {},
  "response_metadata": {},
  "tool_calls": [],
  "tool_call_chunks": [],
  "invalid_tool_calls": []
}
AIMessageChunk {
  "content": "J",
  "additional_kwargs": {},
  "response_metadata": {},
  "tool_calls": [],
  "tool_call_chunks": [],
  "invalid_tool_calls": []
}
AIMessageChunk {
  "content": "'",
  "additional_kwargs": {},
  "response_metadata": {},
  "tool_calls": [],
  "tool_call_chunks": [],
  "invalid_tool_calls": []
}
AIMessageChunk {
  "content": "adore",
  "additional_kwargs": {},
  "response_metadata": {},
  "tool_calls": [],
  "tool_call_chunks": [],
  "invalid_tool_calls": []
}
AIMessageChunk {
  "content": " programmer",
  "additional_kwargs": {},
  "response_metadata": {},
  "tool_calls": [],
  "tool_call_chunks": [],
  "invalid_tool_calls": []
}
AIMessageChunk {
  "content": ".\"",
  "additional_kwargs": {},
  "response_metadata": {},
  "tool_calls": [],
  "tool_call_chunks": [],
  "invalid_tool_calls": []
}
AIMessageChunk {
  "content": "",
  "additional_kwargs": {
    "eventType": "stream-end"
  },
  "response_metadata": {
    "eventType": "stream-end",
    "response_id": "687f94a6-13b7-4c2c-98be-9ca5573c722f",
    "text": "\"J'adore programmer.\"",
    "generationId": "d62c8989-8af5-4357-af79-4ea8e6eb2baa",
    "chatHistory": [
      {
        "role": "USER",
        "message": "Translate \"I love programming\" into French."
      },
      {
        "role": "CHATBOT",
        "message": "\"J'adore programmer.\""
      }
    ],
    "finishReason": "COMPLETE",
    "meta": {
      "apiVersion": {
        "version": "1"
      },
      "billedUnits": {
        "inputTokens": 9,
        "outputTokens": 6
      },
      "tokens": {
        "inputTokens": 75,
        "outputTokens": 6
      }
    }
  },
  "tool_calls": [],
  "tool_call_chunks": [],
  "invalid_tool_calls": [],
  "usage_metadata": {
    "input_tokens": 75,
    "output_tokens": 6,
    "total_tokens": 81
  }
}

Aggregate Streamed Chunks 
import { AIMessageChunk } from '@langchain/core/messages';
import { concat } from '@langchain/core/utils/stream';

const stream = await llm.stream(input);
let full: AIMessageChunk | undefined;
for await (const chunk of stream) {
  full = !full ? chunk : concat(full, chunk);
}
console.log(full);

AIMessageChunk {
  "content": "\"J'adore programmer.\"",
  "additional_kwargs": {
    ...
  },
  "response_metadata": {
    "is_finished": false,
    "generationId": "303e0215-96f4-4da5-8c2a-10da3840afce303e0215-96f4-4da5-8c2a-10da3840afce",
    "response_id": "6a8cb7ef-f1b9-44f6-a1df-67aa506d3f0f",
    "text": "\"J'adore programmer.\"",
    "chatHistory": [
      {
        "role": "USER",
        "message": "Translate \"I love programming\" into French."
      },
      {
        "role": "CHATBOT",
        "message": "\"J'adore programmer.\""
      }
    ],
    "finishReason": "COMPLETE",
    "meta": {
      "apiVersion": {
        "version": "1"
      },
      "billedUnits": {
        "inputTokens": 9,
        "outputTokens": 6
      },
      "tokens": {
        "inputTokens": 75,
        "outputTokens": 6
      }
    }
  },
  "tool_calls": [],
  "tool_call_chunks": [],
  "invalid_tool_calls": [],
  "usage_metadata": {
    "input_tokens": 75,
    "output_tokens": 6,
    "total_tokens": 81
  }
}

Bind tools 
import { z } from 'zod';

const GetWeather = {
  name: "GetWeather",
  description: "Get the current weather in a given location",
  schema: z.object({
    location: z.string().describe("The city and state, e.g. San Francisco, CA")
  }),
}

const GetPopulation = {
  name: "GetPopulation",
  description: "Get the current population in a given location",
  schema: z.object({
    location: z.string().describe("The city and state, e.g. San Francisco, CA")
  }),
}

const llmWithTools = llm.bindTools([GetWeather, GetPopulation]);
const aiMsg = await llmWithTools.invoke(
  "Which city is hotter today and which is bigger: LA or NY?"
);
console.log(aiMsg.tool_calls);

[
  {
    name: 'GetWeather',
    args: { location: 'LA' },
    id: 'ce8076ee-2ed3-429d-938c-14f3218c',
    type: 'tool_call'
  },
  {
    name: 'GetWeather',
    args: { location: 'NY' },
    id: '23d1a96e-3a2c-46f4-9d9e-cccd02c6',
    type: 'tool_call'
  },
  {
    name: 'GetPopulation',
    args: { location: 'LA' },
    id: '2bf9d627-310f-46ff-93a9-86baeae9',
    type: 'tool_call'
  },
  {
    name: 'GetPopulation',
    args: { location: 'NY' },
    id: 'c95e6ac0-ee9b-48de-86b2-12548fd1',
    type: 'tool_call'
  }
]

Structured Output 
import { z } from 'zod';

const Joke = z.object({
  setup: z.string().describe("The setup of the joke"),
  punchline: z.string().describe("The punchline to the joke"),
  rating: z.number().optional().describe("How funny the joke is, from 1 to 10")
}).describe('Joke to tell user.');

const structuredLlm = llm.withStructuredOutput(Joke, { name: "Joke" });
const jokeResult = await structuredLlm.invoke("Tell me a joke about cats");
console.log(jokeResult);

{
  punchline: 'Because she wanted to be a first-aid kit.',
  rating: 5,
  setup: 'Why did the cat join the Red Cross?'
}

Usage Metadata 
const aiMsgForMetadata = await llm.invoke(input);
console.log(aiMsgForMetadata.usage_metadata);

{ input_tokens: 75, output_tokens: 6, total_tokens: 81 }

Response Metadata 
const aiMsgForResponseMetadata = await llm.invoke(input);
console.log(aiMsgForResponseMetadata.response_metadata);

{
  estimatedTokenUsage: { completionTokens: 6, promptTokens: 75, totalTokens: 81 },
  response_id: 'a688ad65-4db2-4a7a-b6aa-124aa2410319',
  generationId: 'ee259727-18c5-43f7-b9bd-a2a60c0c040b',
  chatHistory: [
    {
      role: 'USER',
      message: 'Translate "I love programming" into French.'
    },
    { role: 'CHATBOT', message: `"J'adore programmer."` }
  ],
  finishReason: 'COMPLETE',
  meta: {
    apiVersion: { version: '1' },
    billedUnits: { inputTokens: 9, outputTokens: 6 },
    tokens: { inputTokens: 75, outputTokens: 6 }
  }
}

Type Parameters

Hierarchy (view full)

Implements

Constructors

constructor

Properties

client model streamUsage streaming temperature

Methods

bindTools getLsParams invocationParams

Constructors

constructor

Properties

client

client: CohereClient

model

model: string = "command-r-plus"

The name of the model to use.

Default

{"command"}

streamUsage

streamUsage: boolean = true

Whether or not to include token usage when streaming. This will include an extra chunk at the end of the stream with eventType: "stream-end" and the token usage in usage_metadata.

Default

{true}

streaming

streaming: boolean = false

Whether or not to stream the response.

Default

{false}

temperature

temperature: number = 0.3

What sampling temperature to use, between 0.0 and 2.0. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.

Default

{0.3}

Methods

bindTools

getLsParams

invocationParams

Settings

Member Visibility

On This Page

Constructor argsRuntime argsExamples
Constructors
Properties
Methods