DeepSeek-R1 now obtainable as a totally managed serverless mannequin in Amazon Bedrock

As of January 30, DeepSeek-R1 fashions grew to become obtainable in Amazon Bedrock via the Amazon Bedrock Market and Amazon Bedrock Customized Mannequin Import. Since then, hundreds of consumers have deployed these fashions in Amazon Bedrock. Clients worth the strong guardrails and complete tooling for protected AI deployment. In the present day, we’re making it even simpler to make use of DeepSeek in Amazon Bedrock via an expanded vary of choices, together with a brand new serverless answer.

The totally managed DeepSeek-R1 mannequin is now usually obtainable in Amazon Bedrock. Amazon Internet Providers (AWS) is the primary cloud service supplier (CSP) to ship DeepSeek-R1 as a totally managed, usually obtainable mannequin. You’ll be able to speed up innovation and ship tangible enterprise worth with DeepSeek on AWS with out having to handle infrastructure complexities. You’ll be able to energy your generative AI purposes with DeepSeek-R1’s capabilities utilizing a single API within the Amazon Bedrock’s totally managed service and get the advantage of its in depth options and tooling.

In accordance with DeepSeek, their mannequin is publicly obtainable beneath MIT license and affords robust capabilities in reasoning, coding, and pure language understanding. These capabilities energy clever determination assist, software program growth, mathematical problem-solving, scientific evaluation, knowledge insights, and complete information administration techniques.

As is the case for all AI options, give cautious consideration to knowledge privateness necessities when implementing in your manufacturing environments, examine for bias in output, and monitor your outcomes. When implementing publicly obtainable fashions like DeepSeek-R1, take into account the next:

• Knowledge safety – You’ll be able to entry the enterprise-grade safety, monitoring, and price management options of Amazon Bedrock which can be important for deploying AI responsibly at scale, all whereas retaining full management over your knowledge. Customers’ inputs and mannequin outputs aren’t shared with any mannequin suppliers. You need to use these key security measures by default, together with knowledge encryption at relaxation and in transit, fine-grained entry controls, safe connectivity choices, and obtain numerous compliance certifications whereas speaking with the DeepSeek-R1 mannequin in Amazon Bedrock.
• Accountable AI – You’ll be able to implement safeguards personalized to your utility necessities and accountable AI insurance policies with Amazon Bedrock Guardrails. This consists of key options of content material filtering, delicate data filtering, and customizable safety controls to stop hallucinations utilizing contextual grounding and Automated Reasoning checks. This implies you’ll be able to management the interplay between customers and the DeepSeek-R1 mannequin in Bedrock together with your outlined set of insurance policies by filtering undesirable and dangerous content material in your generative AI purposes.
• Mannequin analysis – You’ll be able to consider and evaluate fashions to determine the optimum mannequin to your use case, together with DeepSeek-R1, in just a few steps via both computerized or human evaluations through the use of Amazon Bedrock mannequin analysis instruments. You’ll be able to select computerized analysis with predefined metrics reminiscent of accuracy, robustness, and toxicity. Alternatively, you’ll be able to select human analysis workflows for subjective or customized metrics reminiscent of relevance, model, and alignment to model voice. Mannequin analysis gives built-in curated datasets, or you’ll be able to herald your individual datasets.

We strongly suggest integrating Amazon Bedrock Guardrails and utilizing Amazon Bedrock mannequin analysis options together with your DeepSeek-R1 mannequin so as to add strong safety to your generative AI purposes. To be taught extra, go to Shield your DeepSeek mannequin deployments with Amazon Bedrock Guardrails and Consider the efficiency of Amazon Bedrock assets.

Get began with the DeepSeek-R1 mannequin in Amazon Bedrock
In case you’re new to utilizing DeepSeek-R1 fashions, go to the Amazon Bedrock console, select Mannequin entry beneath Bedrock configurations within the left navigation pane. To entry the totally managed DeepSeek-R1 mannequin, request entry for DeepSeek-R1 in DeepSeek. You’ll then be granted entry to the mannequin in Amazon Bedrock.

Subsequent, to check the DeepSeek-R1 mannequin in Amazon Bedrock, select Chat/Textual content beneath Playgrounds within the left menu pane. Then select Choose mannequin within the higher left, and choose DeepSeek because the class and DeepSeek-R1 because the mannequin. Then select Apply.

Utilizing the chosen DeepSeek-R1 mannequin, I run the next immediate instance:

A household has $5,000 to save lots of for his or her trip subsequent yr. They will place the cash in a financial savings account incomes 2% curiosity yearly or in a certificates of deposit incomes 4% curiosity yearly however with no entry to the funds till the holiday. In the event that they want $1,000 for emergency bills in the course of the yr, how ought to they divide their cash between the 2 choices to maximise their trip fund?

This immediate requires a posh chain of thought and produces very exact reasoning outcomes.

To be taught extra about utilization suggestions for prompts, consult with the DeepSeek-R1 mannequin immediate information.

By selecting View API request, you can too entry the mannequin utilizing code examples within the AWS Command Line Interface (AWS CLI) and AWS SDK. You need to use us.deepseek.r1-v1:0 because the mannequin ID.

Here’s a pattern of the AWS CLI command:

aws bedrock-runtime invoke-model 
       --model-id us.deepseek.r1-v1:0 
       --body "{"immediate": "<｜begin_of_sentence｜><｜Person｜>Type_Your_Prompt_Here<｜Assistant｜>n", "max_tokens": 512, "temperature": 0.5, "top_p": 0.9}" 
       --cli-binary-format raw-in-base64-out 
       --region us-west-2 
       invoke-model-output.txt

The mannequin helps each the InvokeModel and Converse API. The next Python code examples present easy methods to ship a textual content message to the DeepSeek-R1 mannequin utilizing the Amazon Bedrock Converse API for textual content technology. To be taught extra, go to DeepSeek mannequin inference parameters and responses within the AWS documentation.

import boto3
from botocore.exceptions import ClientError

# Create a Bedrock Runtime consumer within the AWS Area you need to use.
consumer = boto3.consumer("bedrock-runtime", region_name="us-west-2")

# Set the mannequin ID, e.g., DeepSeek-R1 Mannequin.
model_id = "us.deepseek.r1-v1:0"

# Begin a dialog with the person message.
user_message = "Type_Your_Prompt_Here"
dialog = [
    {
        "role": "user",
        "content": [{"text": user_message}],
    }
]

strive:
    # Ship the message to the mannequin, utilizing a fundamental inference configuration.
    response = consumer.converse(
        modelId=model_id,
        messages=dialog,
        inferenceConfig={"maxTokens": 512, "temperature": 0.5, "topP": 0.9},
    )

    # Extract and print the response textual content.
    response_text = response["output"]["message"]["content"][0]["text"]
    print(response_text)

besides (ClientError, Exception) as e:
    print(f"ERROR: Cannot invoke '{model_id}'. Cause: {e}")
    exit(1)

To allow Amazon Bedrock Guardrails on the DeepSeek-R1 mannequin, choose Guardrails beneath Safeguards within the left navigation pane, and create a guardrail by configuring as many filters as you want. For instance, for those who filter for “politics” phrase, your guardrails will acknowledge this phrase within the immediate and present you the blocked message.

You’ll be able to check the guardrail with completely different inputs to evaluate the guardrail’s efficiency. You’ll be able to refine the guardrail by setting denied subjects, phrase filters, delicate data filters, and blocked messaging till it matches your wants.

To be taught extra about Amazon Bedrock Guardrails, go to Cease dangerous content material in fashions utilizing Amazon Bedrock Guardrails within the AWS documentation or different deep dive weblog posts about Amazon Bedrock Guardrails on the AWS Machine Studying Weblog channel.

Right here’s a demo walkthrough highlighting how one can reap the benefits of the totally managed DeepSeek-R1 mannequin in Amazon Bedrock:

Now obtainable
DeepSeek-R1 is now obtainable totally managed in Amazon Bedrock within the US East (N. Virginia), US East (Ohio), and US West (Oregon) AWS Areas via cross-Area inference. Examine the full Area record for future updates. To be taught extra, take a look at the DeepSeek in Amazon Bedrock product web page and the Amazon Bedrock pricing web page.

Give the DeepSeek-R1 mannequin a strive within the Amazon Bedrock console in the present day and ship suggestions to AWS re:Submit for Amazon Bedrock or via your normal AWS Assist contacts.

— Channy

Up to date on March 10, 2025 — Mounted screenshots of mannequin choice and mannequin ID.

Up to date on March 13, 2025 — Added information hyperlinks of DeepSeek-R1 mannequin prompts and mannequin inference parameters and responses.