Run scalable, enterprise-grade generative AI workloads with Cohere Command R & R+, now obtainable in Amazon Bedrock

In November 2023, we made two new Cohere fashions obtainable in Amazon Bedrock (Cohere Command Gentle and Cohere Embed English). In the present day, we’re asserting the addition of two extra Cohere fashions in Amazon Bedrock; Cohere Command R and Command R+.

Organizations want generative synthetic intelligence (generative AI) fashions to securely work together with data saved of their enterprise knowledge sources. Each Command R and Command R+ are highly effective, scalable giant language fashions (LLMs), purpose-built for real-world, enterprise-grade workloads. These fashions are multilingual and are centered on balancing excessive effectivity with robust accuracy to excel at capabilities equivalent to Retrieval-Augmented Technology (RAG), and gear use to allow enterprises to maneuver past proof-of-concept (POC), and into manufacturing utilizing synthetic intelligence (AI).

Command R is a scalable multilingual generative mannequin focusing on RAG and gear use to allow production-scale AI for enterprises. Command R+ is a state-of-the-art RAG-optimized mannequin designed to deal with enterprise-grade workloads and optimize enterprise AI purposes. Command R+ is optimized for superior RAG to supply enterprise-ready, extremely dependable, and verifiable responses as a consequence of in-line citations which come normal with this mannequin. With these new Cohere fashions in Bedrock, you possibly can scale with AI to rapidly find probably the most related data to assist duties throughout enterprise capabilities like finance, human sources (HR), gross sales, advertising, and buyer assist, amongst others, in a variety of enterprise sectors. Software use can also be obtainable in Command R+. Command R+ is a robust multilingual mannequin, and equally to Command R, contains a tokenizer that compresses non-English textual content a lot better than the tokenizer used for different fashions available in the market.

Getting began with Command R and Command R+
To get began with each fashions in Amazon Bedrock, you first have to get entry to the fashions. Within the Amazon Bedrock console, select Mannequin entry, after which select Handle mannequin entry. Subsequent, select your most well-liked mannequin(s), after which select Save modifications. As you possibly can see, you now have six Cohere fashions to select from in Amazon Bedrock – together with Command R and Command R+ – providing you with higher selection and adaptability to make use of the optimum fashions to your particular enterprise wants.

After getting entry to your most well-liked mannequin, you need to use the mannequin in Amazon Bedrock. Refresh the bottom fashions desk to view the up to date standing.

The fashions have been educated to reply within the language of the consumer equivalent to English, French, Spanish, Italian, German, Brazilian Portuguese, Japanese, Korean, Simplified Chinese language, and Arabic. Right here’s an instance:

Immediate

<s>"Écris une description de produit pour une voiture électrique en 50 à 75 mots"

Output

Découvrez la voiture électrique qui va révolutionner votre façon de conduire.
Avec son design élégant, cette voiture offre une expérience de conduite distinctive
avec une accélération puissante et une autonomie impressionnante. Sa
technologie avancée vous garantit une cost rapide et une fiabilité inégalée.
Avec sa conception innovante et sturdy, cette voiture est parfaite pour les 
trajets urbains et les longues distances. Profitez d'une conduite silencieuse
et vivez l'expérience de la voiture électrique!

Programmatically work together with Command R and Command R+
You may also use AWS Command Line Interface (CLI) and AWS Software program Improvement Equipment (SDK) to make varied calls utilizing Amazon Bedrock APIs. Following, is a pattern code in Python that interacts with Amazon Bedrock Runtime APIs with AWS SDK. Taking the identical textual content era immediate I used earlier, right here is the way it seems when used programmatically. On this instance I’m interacting with the Command R mannequin. Again to Python, I first run the ListFoundationModels API name to find the modelId for Command R.

import boto3
import json
import numpy

bedrock = boto3.consumer(service_name="bedrock", region_name="us-east-1")

listModels = bedrock.list_foundation_models(byProvider="cohere")
print("n".be part of(checklist(map(lambda x: f"{x['modelName']} : { x['modelId'] }", listModels['modelSummaries']))))

Working this code offers the checklist:

Command : cohere.command-text-v14
Command Gentle : cohere.command-light-text-v14
Embed English : cohere.embed-english-v3
Embed Multilingual : cohere.embed-multilingual-v3
Command R: cohere.command-r-v1:0
Command R+: cohere.command-r-plus-v1:0

From this checklist, I choose cohere.command-r-v1:0 mannequin ID and write the code to generate the textual content as proven earlier on this publish.

import boto3
import json

bedrock = boto3.consumer(service_name="bedrock-runtime", region_name="us-east-1")

immediate = """
<s>Écris une description de produit pour une voiture électrique en 50 à 75 mots

physique = json.dumps({
    "immediate": immediate,
    "max_tokens": 512,
    "top_p": 0.8,
    "temperature": 0.5,
})

modelId = "cohere.command-r-v1:0"

settle for = "utility/json"
contentType = "utility/json"

response = bedrock.invoke_model(
    physique=physique,
    modelId=modelId,
    settle for=settle for,
    contentType=contentType
)

print(json.masses(response.get('physique').learn()))

You will get JSON formatted output as like:

Découvrez la voiture électrique qui va révolutionner votre façon de conduire.
Avec son design élégant, cette voiture offre une expérience de conduite distinctive
avec une accélération puissante et une autonomie impressionnante. Sa
technologie avancée vous garantit une cost rapide et une fiabilité inégalée.
Avec sa conception innovante et sturdy, cette voiture est parfaite pour les 
trajets urbains et les longues distances. Profitez d'une conduite silencieuse
et vivez l'expérience de la voiture électrique!

Now Accessible

Command R and Command R+ fashions, together with different Cohere fashions, can be found at present in Amazon Bedrock within the US East (N. Virginia) and US West (Oregon) Areas; test the full Area checklist for future updates.

Go to our group.aws website to search out deep-dive technical content material and to find how our Builder communities are utilizing Amazon Bedrock of their options. Give Command R and Command R+ a attempt within the Amazon Bedrock console at present and ship suggestions to AWS re:Submit for Amazon Bedrock or by means of your standard AWS Assist contacts.

– Veliswa.