How do I get a Cohere API Key?

Log in to the [**Cohere Dashboard**](https://dashboard.cohere.com/api-keys), go to **API Keys** and click **Create API Key**. Copy the key immediately — it starts with a random string and won't be shown again. Free tier includes trial access with rate limits.

What models are available?

Use the `list_models` tool to see all available Cohere models. Key models include command-r-plus (most capable, 128K context), command-r (efficient, 128K context), command-r7b (lightweight, 128K context), embed-v4 (embeddings) and rerank-v3.5 (reranking).

Can I send multi-turn conversations?

Yes! Pass a messages array with alternating 'user', 'assistant' and 'system' roles. Each message has a 'role' and 'content' field. Command models support function calling and will return tool_calls when appropriate.

What is reranking and when should I use it?

Reranking reorders a set of documents by their relevance to a query. Use it after an initial search to improve result quality. The rerank tool takes a query, list of documents and returns them ranked by relevance score. Cohere's rerank models are industry-leading for search applications.

Cohere MCP Server for LangChain 6 tools — connect in under 2 minutes

Name: Cohere
Availability: InStock
Author: Vinkius

python.langchain.com/docs/integrations/tools/mcp

Built by Vinkius GDPR 6 Tools Framework

LangChain is the leading Python framework for composable LLM applications. Connect Cohere through the Vinkius and LangChain agents can call every tool natively — combine them with retrievers, memory, and output parsers for sophisticated AI pipelines.

Get MCP Server for AI Agents

ASK AI ABOUT THIS MCP SERVER

Open in ChatGPT Open in Claude Open in Perplexity

Vinkius supports streamable HTTP and SSE.

python

import asyncio
from langchain_mcp_adapters.client import MultiServerMCPClient
from langchain_openai import ChatOpenAI
from langgraph.prebuilt import create_react_agent

async def main():
    # Your Vinkius token — get it at cloud.vinkius.com
    async with MultiServerMCPClient({
        "cohere": {
            "transport": "streamable_http",
            "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp",
        }
    }) as client:
        tools = client.get_tools()
        agent = create_react_agent(
            ChatOpenAI(model="gpt-4o"),
            tools,
        )
        response = await agent.ainvoke({
            "messages": [{
                "role": "user",
                "content": "Using Cohere, show me what tools are available.",
            }]
        })
        print(response["messages"][-1].content)

asyncio.run(main())

Fully ManagedVinkius Servers

60%Token savings

High SecurityEnterprise-grade

IAMAccess control

EU AI ActCompliant

DLPData protection

V8 IsolateSandboxed

Ed25519Audit chain

<40msKill switch

Stream every event to Splunk, Datadog, or your own webhook in real-time

* Every MCP server runs on Vinkius-managed infrastructure inside AWS - a purpose-built runtime with per-request V8 isolates, Ed25519 signed audit chains, and sub-40ms cold starts optimized for native MCP execution. See our infrastructure

About Cohere MCP Server

Connect your Cohere account to any AI agent and leverage enterprise-grade AI models through natural conversation.

LangChain's ecosystem of 500+ components combines seamlessly with Cohere through native MCP adapters. Connect 6 tools via the Vinkius and use ReAct agents, Plan-and-Execute strategies, or custom agent architectures — with LangSmith tracing giving full visibility into every tool call, latency, and token cost.

What you can do

Model Discovery — List all available Cohere models with their names, capabilities and context lengths
Chat API — Send conversations to Command models (command-r-plus, command-r, command-r7b) and receive responses with citations and tool call support
Embeddings — Generate vector embeddings for semantic search with multiple embedding types (float, int8, uint8, binary)
Reranking — Rerank documents by relevance to a search query using Cohere's industry-leading reranking models
Tokenization — Tokenize and detokenize text for estimating token counts and debugging

The Cohere MCP Server exposes 6 tools through the Vinkius. Connect it to LangChain in under two minutes — no API keys to rotate, no infrastructure to provision, no vendor lock-in. Your configuration, your data, your control.

How to Connect Cohere to LangChain via MCP

Follow these steps to integrate the Cohere MCP Server with LangChain.

Install dependencies

Run pip install langchain langchain-mcp-adapters langgraph langchain-openai

Replace the token

Replace [YOUR_TOKEN_HERE] with your Vinkius token

Run the agent

Save the code and run python agent.py

Explore tools

The agent discovers 6 tools from Cohere via MCP

Why Use LangChain with the Cohere MCP Server

LangChain provides unique advantages when paired with Cohere through the Model Context Protocol.

The largest ecosystem of integrations, chains, and agents — combine Cohere MCP tools with 500+ LangChain components

Agent architecture supports ReAct, Plan-and-Execute, and custom strategies with full MCP tool access at every step

LangSmith tracing gives you complete visibility into tool calls, latencies, and token usage for production debugging

Memory and conversation persistence let agents maintain context across Cohere queries for multi-turn workflows

Cohere + LangChain Use Cases

Practical scenarios where LangChain combined with the Cohere MCP Server delivers measurable value.

RAG with live data: combine Cohere tool results with vector store retrievals for answers grounded in both real-time and historical data

Autonomous research agents: LangChain agents query Cohere, synthesize findings, and generate comprehensive research reports

Multi-tool orchestration: chain Cohere tools with web scrapers, databases, and calculators in a single agent run

Production monitoring: use LangSmith to trace every Cohere tool call, measure latency, and optimize your agent's performance

Cohere MCP Tools for LangChain (6)

These 6 tools become available when you connect Cohere to LangChain via MCP:

chat

Requires the model ID (e.g. "command-r-plus", "command-r", "command-r7b") and messages array in JSON format. Each message must have a "role" ("user", "assistant", "system" or "tool") and "content" (text or array of content blocks). Optionally set max_tokens, temperature (0-1), p (nucleus sampling 0-1) and tools array for function calling. Returns the model's response with text, citations and tool calls. Send a chat message to a Cohere model

detokenize

Requires the token IDs array. Returns the reconstructed text. Useful for debugging and verifying tokenization. Detokenize token IDs back to text using Cohere

embed

Requires the model ID (e.g. "embed-v4", "embed-v3"), texts array and input_type ("search_document", "search_query", "classification", "clustering"). Returns embedding vectors for each input text. Useful for semantic search, similarity comparison and vector database storage. Generate embeddings using Cohere

list_models

Each model returns its name (e.g. "command-r-plus", "command-r", "embed-v4", "rerank-v3.5"), endpoint compatibility, context length and tokenization info. Use this to discover which models are available and their capabilities. List all available Cohere models

rerank

Requires the model ID (e.g. "rerank-v3.5", "rerank-english-v3.0"), query text and documents array. Optionally set top_n to return only the top N results. Returns ranked documents with relevance scores. Rerank documents by relevance to a query

tokenize

Requires the text to tokenize and optionally the model. Returns the list of token IDs and token strings. Useful for estimating token counts before sending to chat or embed endpoints. Tokenize text using Cohere

Example Prompts for Cohere in LangChain

Ready-to-use prompts you can give your LangChain agent to start working with Cohere immediately.

"Send a message to Command R+ asking 'What is the capital of Brazil?'"

"Rerank these documents for the query 'machine learning models': ['Neural networks are inspired by biological neurons.', 'Python is a popular programming language.', 'Transformers use attention mechanisms for sequence processing.']"

"Generate embeddings for these texts: ['The weather is nice today.', 'I love programming in Python.'] using embed-v4."

Troubleshooting Cohere MCP Server with LangChain

Common issues when connecting Cohere to LangChain through the Vinkius, and how to resolve them.

MultiServerMCPClient not found

Install: pip install langchain-mcp-adapters

Cohere + LangChain FAQ

Common questions about integrating Cohere MCP Server with LangChain.

How does LangChain connect to MCP servers?

Use langchain-mcp-adapters to create an MCP client. LangChain discovers all tools and wraps them as native LangChain tools compatible with any agent type.

Which LangChain agent types work with MCP?

All agent types including ReAct, OpenAI Functions, and custom agents work with MCP tools. The tools appear as standard LangChain tools after the adapter wraps them.