This feature is only available on the Enterprise tier
Ona Agent supports integration with AWS Bedrock. This guide shows you how to configure this integration.

​
Prerequisites

  • You must have an AWS account with Bedrock enabled and model access approved.
  • You must have an Ona enterprise license.
  • Your enterprise runner must be deployed in the same AWS account and have IAM permissions to invoke Bedrock.

​
Set up AWS permissions and model access

  1. Go to the AWS Bedrock console
  2. Open Model access and request access to the Anthropic Claude model(s) you plan to use. We recommend using Claude Sonnet 4.
  3. Wait for approval

​
Configure the Bedrock endpoint

​
Endpoint format

bedrock://<model-id>
Find model IDs in the AWS Bedrock model names. Use Claude Sonnet 4 model identifiers for your region.

​
Add the configuration to Ona

You can configure a Bedrock integration in two ways:
  • Gitpod CLI
  • Through the UI

​
Option 1: Gitpod CLI

gitpod runner config llm-integration create <runner-id> SUPPORTED_MODEL_SONNET_4 bedrock://<model-id> ignore

# Optional: set a specific maximum token limit
# We generally do not recommend setting a maximum token limit; prefer defaults unless you have specific constraints.
gitpod runner config llm-integration set-max-tokens <runner-id> SUPPORTED_MODEL_SONNET_4 4000

# Verify integrations
gitpod runner config llm-integration list

​
Option 2: Through the UI

  1. Go to the Runners settings page
  2. Select an enterprise runner where you want to enable LLM integration
  3. Scroll down to the “LLM Providers” section
  4. Click the Configure button

LLM Provider configuration screen

  1. In “Bedrock Model ID”, enter the model ID for your region (for example, eu.anthropic.claude-sonnet-4-20250514-v1:0). Tip: Verify the identifier for your region in the AWS Bedrock model IDs documentation.
  2. “API Key”: Not required for Bedrock (AWS credentials are used).
  3. If a model preset selector is shown, choose “Claude 4 Sonnet”.
  4. Click Create Integration

LLM Provider configuration screen

​
Verify the integration

  1. Create a new environment with the configured runner
  2. Open Ona Agent and start a conversation
  3. If Ona Agent responds, the integration is working. Test with a simple code generation request

​
Supported models

AWS Bedrock supports Anthropic Claude models:
  • Claude Sonnet 4 (Highly recommended)
  • Claude Sonnet 3.7 (Supported but not recommended - only use when Sonnet 4 is not available in your region)

​
Identifying available models

There are two ways to identify models: foundation models and inference profiles. Inference profiles are resources that define a model and one or more regions for routing requests, enabling cross-region inference, usage tracking, and cost monitoring. Availability varies by region — some regions only support foundation models, while others support both foundation models and inference profiles. Check model availability per region in the AWS documentation. You can check availability using AWS CLI commands from your environment, assuming you have proper authentication and region environment variables set: 
# List available inference profiles
aws bedrock list-inference-profiles

# List available foundation models
aws bedrock list-foundation-models

​
Testing model connectivity

To run a simple smoke test and verify a model works: 
# Set your foundation model or inference profile ID
export MODEL_ID="eu.anthropic.claude-sonnet-4-20250514-v1:0"

# Test the model with a simple request
aws bedrock-runtime invoke-model \
    --model-id "$MODEL_ID" \
    --body '{
        "messages": [
            {
                "role": "user",
                "content": "Hi there!"
            }
        ],
        "max_tokens": 50,
        "anthropic_version": "bedrock-2023-05-31"
    }' \
    --cli-binary-format raw-in-base64-out
If the smoke test succeeds, the model should work with Ona Agent.

​
Troubleshooting

​
Common issues

Model not available in region
  • Use a model ID that is available in your runner’s region
  • For regionalized models, use the correct prefix (for example, us.anthropic.*)
AccessDeniedException when invoking model
  • Ensure the runner role has bedrock:InvokeModel and bedrock:InvokeModelWithResponseStream
  • Confirm model access is approved in the Bedrock console
Invalid model URI format
  • Use bedrock://<model-id> and include the full version (for example, ...-v1:0)
Throttling or token limit errors
  • Lower the maximum tokens with gitpod runner config llm-integration set-max-tokens ...
  • Review Bedrock service quotas and request increases if needed

​
Getting help

If you encounter issues:
  1. Check Ona Agent logs for detailed error messages
  2. Verify AWS quotas and Bedrock model access
  3. Contact your account manager for additional support

​
Next steps