Skip to main content

Cohere

Setup​

  1. You can get an API key from Cohere.
  2. The API key can be configured as an environment variable (COHERE_API_KEY) or passed in as an option into the model constructor.

Model Functions​

Examples

Generate Text​

CohereTextGenerationModel API

import { cohere, generateText } from "modelfusion";

const text = await generateText({
  model: cohere.TextGenerator({
    model: "command",
    temperature: 0.7,
    maxGenerationTokens: 500,
  }),
  prompt: "Write a short story about a robot learning to love:\n\n",
});

Stream Text​

CohereTextGenerationModel API

import { cohere, streamText } from "modelfusion";

const textStream = await streamText({
  model: cohere.TextGenerator({
    model: "command",
    temperature: 0.7,
    maxGenerationTokens: 500,
  }),
  prompt: "Write a short story about a robot learning to love:\n\n",
});

for await (const textPart of textStream) {
  process.stdout.write(textPart);
}

Embed Text​

CohereTextEmbeddingModel API

import { cohere, embedMany } from "modelfusion";

const embeddings = await embedMany({
  model: cohere.TextEmbedder({ model: "embed-english-light-v2.0" }),
  values: [
    "At first, Nox didn't know what to do with the pup.",
    "He keenly observed and absorbed everything around him, from the birds in the sky to the trees in the forest.",
  ],
});

Tokenize Text​

CohereTokenizer API

import { cohere, countTokens } from "modelfusion";

const tokenizer = cohere.Tokenizer({ model: "command" });

const text = "At first, Nox didn't know what to do with the pup.";

const tokenCount = await countTokens(tokenizer, text);
const tokens = await tokenizer.tokenize(text);
const tokensAndTokenTexts = await tokenizer.tokenizeWithTexts(text);
const reconstructedText = await tokenizer.detokenize(tokens);

Configuration​

API Configuration​

Cohere API Configuration

const api = cohere.Api({
  apiKey: "my-api-key",
  // ...
});

const model = cohere.TextGenerator({
  api,
  // ...
});