Cohere Sdk

Cohere SDK Skill

[!CRITICAL] MANDATORY DIRECTIVE: DEFER TO DOCUMENTED MODEL CATALOG DO NOT rely on internal training knowledge for Cohere model selection. ALWAYS consult the active catalog: Cohere Models Catalog. The current production standard is led by Command A+ (command-a-plus-05-2026) and Command A (command-a-03-2025). The v1 CohereClient chat API is legacy; use CohereClientV2.

1. SDK Availability & Setup

Package Installation

# Node.js / TypeScript
npm install cohere-ai

# Python
pip install cohere

Client Initialization

TypeScript / JavaScript

import { CohereClientV2 } from 'cohere-ai';

const client = new CohereClientV2({
  token: process.env.COHERE_API_KEY,
});

Python

import os
import cohere

client = cohere.ClientV2(api_key=os.environ.get("COHERE_API_KEY"))

[!WARNING] Do not use CohereClient / cohere.Client for new chat, embed, or rerank calls. Those are the v1 APIs.

2. Core Implementation Patterns

2.1 Chat (client.chat)

import { CohereClientV2 } from 'cohere-ai';
const client = new CohereClientV2({ token: process.env.COHERE_API_KEY });

const response = await client.chat({
  model: 'command-a-plus-05-2026',
  messages: [
    { role: 'system', content: 'You are an expert distributed systems engineer.' },
    { role: 'user', content: 'Explain Paxos consensus in 3 concise bullet points.' },
  ],
  temperature: 0.2,
});

const text = response.message.content
  ?.filter((block) => block.type === 'text')
  .map((block) => block.text)
  .join('') ?? '';
console.log(text);

2.2 Streaming Chat (client.chatStream)

const stream = await client.chatStream({
  model: 'command-a-03-2025',
  messages: [{ role: 'user', content: 'Write a haiku on compiler optimization.' }],
});

for await (const event of stream) {
  if (event.type === 'content-delta') {
    const delta = event.delta?.message?.content?.text;
    if (typeof delta === 'string') {
      process.stdout.write(delta);
    }
  }
}
console.log();

2.3 Embeddings (embed-v4.0)

const embed = await client.embed({
  model: 'embed-v4.0',
  inputType: 'search_document',
  texts: ['Document chunk for vector search', 'Search query text'],
  embeddingTypes: ['float'],
});

console.log(embed.embeddings.float?.[0]?.length);

2.4 Rerank (rerank-v4.0-pro)

const ranked = await client.rerank({
  model: 'rerank-v4.0-pro',
  query: 'motion event after hours at the loading dock',
  documents: [
    'Camera 12 saw a person at 02:14 near dock door B.',
    'HVAC temperature drifted 2C in zone 3.',
  ],
  topN: 2,
});

console.log(ranked.results);

3. References & Catalogs

• Cohere Models Catalog
• TypeScript Reference Implementation
• Python Reference Implementation