Use Amazon Bedrock To Try Claude, OpenAI, DeepSeek, And More

Beginner Guide:

This guide is based on the transcript, but with one important correction:

AWS Bedrock is not an “unlimited $20,000 subscription for free.”

What you get is access to model testing through AWS, plus Free Tier credits if your account is eligible. AWS currently says new customers can get up to $200 in credits and use the Free Plan for up to 6 months. Usage beyond credits or free limits can create charges, so set billing alerts before testing.

What Amazon Bedrock Is

Amazon Bedrock is AWS’s platform for using AI models from different providers in one place.

Inside Bedrock, you can test models from providers like:

• Anthropic Claude
• OpenAI
• DeepSeek
• Meta
• Mistral
• Amazon Nova
• Cohere
• Stability AI
• and more, depending on region and access

You can test models in the Playground, or generate API keys for apps, automations, AI agents, and internal tools.

Before You Start

You need:

1. An AWS account
2. A valid email address
3. A payment method
4. Access to Amazon Bedrock in your AWS region
5. Basic billing safety setup

Important: Bedrock model availability changes by region. If you do not see a model, switch region or check the model catalog.

Step 1: Create A Free AWS Account

Go to AWS and create a free account:

https://aws.amazon.com/free

During signup:

1. Enter your email
2. Verify your email
3. Add payment details
4. Complete identity verification
5. Choose the Free Plan if available

For India users, AWS may ask for payment verification through UPI or card. A small verification charge may appear and usually gets refunded.

Step 2: Set Up Billing Safety First

Do this before opening Bedrock.

Inside AWS Console:

1. Search for “Billing”
2. Open “Budgets”
3. Create a monthly budget
4. Set a low limit, for example $1 or $5
5. Add email alerts at 50 percent, 80 percent, and 100 percent

This is boring, but important.

Free credits are useful.

Surprise AWS bills are not.

Step 3: Open Amazon Bedrock

Now go to the AWS Console.

Search:

Amazon Bedrock

Open the Bedrock service.

If Bedrock is not visible, try a supported region like:

• US East, N. Virginia
• US West, Oregon
• Asia Pacific, Mumbai
• Europe, Ireland

Availability depends on model and region.

Step 4: Open The Model Catalog

Inside Bedrock, look for:

Model Catalog

This is where you see available models.

You may find models like:

• Claude Sonnet
• Claude Opus
• DeepSeek
• OpenAI models
• Amazon Nova
• Meta Llama
• Mistral
• Stability AI

The transcript says “Cloud Opus 4.7” and “Sonnet 4.6.”

In Bedrock, look for the actual model names listed in the catalog. Model names and versions change over time.

Step 5: Request Or Enable Model Access

Some models work directly.

Some models need access first.

For Anthropic Claude models, AWS may ask you to submit first-time use details before using the model.

You may need to fill:

• use case
• company or personal use details
• intended application
• responsible AI agreement

After submitting, access may be granted quickly, but sometimes it takes a few minutes.

Step 6: Use The Playground

Once access is ready, open:

Playground

Then choose:

1. Chat or text playground
2. Select model
3. Choose provider
4. Pick the model
5. Enter your prompt
6. Click Run

This lets you test premium models without setting up code.

Use the Playground for:

• prompt testing
• model comparison
• writing tasks
• research tasks
• document understanding
• coding help
• brainstorming
• AI tool testing

AWS says running a prompt in Playground is equivalent to using the model API, so usage can count toward billing or credits.

Step 7: Generate Amazon Bedrock API Keys

If you want to connect Bedrock to your own apps, go to:

API keys

You will usually see two options:

1. Short-term API key
2. Long-term API key

For beginners, long-term keys are easier for testing.

AWS recommends long-term API keys only for exploration and development. For production, use short-term credentials or IAM roles.

Basic flow:

1. Open Amazon Bedrock
2. Go to API keys
3. Choose Long-term API keys
4. Set expiration, for example 30 days
5. Generate key
6. Copy the key safely

Do not share this key publicly.

Do not paste it into random websites.

Do not upload it to GitHub.

Step 8: What You Can Build With The API Key

Once you have a Bedrock API key, you can connect models to:

• AI agents
• n8n automations
• internal tools
• chatbots
• document analyzers
• customer support workflows
• lead qualification systems
• content generation tools
• research assistants

Beginner examples:

1. AI email assistant Take customer emails and generate draft replies.
2. Lead scoring workflow Send lead data to Claude or OpenAI through Bedrock and score the lead.
3. Document summarizer Upload long text and get summaries or action items.
4. n8n automation Connect a form to Bedrock, generate a response, then send it to Slack or Gmail.
5. Internal chatbot Use a model to answer questions from company docs.

Step 9: Beginner Prompt To Test In Playground

Use this first:

You are an AI assistant helping me understand Amazon Bedrock.

Explain this in simple language:
1. What Amazon Bedrock is
2. How the Playground works
3. How API keys work
4. What I should be careful about as a beginner

Then test a real use case:

Act as a sales automation assistant.

Here is a new lead:
Name:
Company:
Role:
Problem:
Budget:

Write:
1. Lead summary
2. Priority score from 1 to 10
3. Best follow-up message
4. Next action for the sales team

Step 10: Best Beginner Use Cases

Start with simple use cases.

Do not begin with complex apps.

Try:

• summarizing documents
• comparing models
• generating sales emails
• classifying support tickets
• creating onboarding checklists
• extracting action items from notes
• building n8n workflows with AI steps
• testing prompts across different models

Step 11: Safety Checklist

Before using Bedrock seriously:

• Set AWS Budget alerts
• Check Free Tier credits
• Confirm model pricing
• Use API keys only for testing
• Set key expiry dates
• Delete keys you no longer use
• Avoid sharing private data
• Do not paste secrets into prompts
• Check region before testing
• Monitor usage inside Billing

Simple Summary

Here is the full beginner path:

1. Create AWS account
2. Set billing alerts
3. Open Amazon Bedrock
4. Go to Model Catalog
5. Choose Claude, OpenAI, DeepSeek, or another model
6. Complete model access steps if needed
7. Open Playground
8. Test prompts
9. Generate API key if you want app access
10. Use it for AI agents, n8n automations, and internal tools

The real value is not “free Claude.”

The real value is learning how AI models work inside cloud infrastructure.

That is how builders move from using chatbots to building AI systems.

Official references:

• AWS Free Tier: https://aws.amazon.com/free/
• Amazon Bedrock pricing: https://aws.amazon.com/bedrock/pricing/
• Amazon Bedrock model access: https://docs.aws.amazon.com/bedrock/latest/userguide/model-access.html
• Amazon Bedrock Playground: https://docs.aws.amazon.com/bedrock/latest/userguide/playgrounds.html
• Amazon Bedrock API keys: https://docs.aws.amazon.com/bedrock/latest/userguide/api-keys.html
• Generate Bedrock API keys: https://docs.aws.amazon.com/bedrock/latest/userguide/api-keys-generate.html