1.7.0

This release adds support for the newly launched Service Tier.

Amazon Bedrock offers three service tiers for model inference: Priority, Standard, and Flex. The Priority tier delivers the fastest response times for a price premium over standard on-demand pricing. It is best suited for mission-critical applications like customer-facing chatbots and real-time language translation services. The Standard tier provides consistent performance for everyday AI tasks, ideal for content generation, text analysis, and routine document processing. For workloads that can handle longer processing times, the Flex tier offers cost-effective processing for a pricing discount, perfect for model evaluations, content summarization, agentic workflows.

New API

Converse

let bedrock = try await BedrockService(
    region: .uswest2,
)

var builder = try ConverseRequestBuilder(with: .openai_gpt_oss_20b)
    .withPrompt("Who are you?")
    .withServiceTier(.priority)

var reply = try await bedrock.converse(with: builder)

InvokeModel (text)

let bedrock = try await BedrockService(
    region: .uswest2,
)

let textCompletion = try await bedrock.completeText(
    "Who are you?",
    with: .openai_gpt_oss_20b,
    serviceTier: .default
)

Changes

• Added support for AWS Bedrock service tier selection (.default, .priority, .flex) for both Converse and InvokeModel APIs
• New ServiceTier parameter available in converse() and completeText() methods
• New withServiceTier() builder method in ConverseRequestBuilder
• Enables control over request prioritization and cost optimization

References

Doc:
https://docs.aws.amazon.com/bedrock/latest/userguide/service-tiers-inference.html

Launch:
https://aws.amazon.com/blogs/aws/new-amazon-bedrock-service-tiers-help-you-match-ai-workload-performance-with-cost/

What's Changed (automatically generated)

• Move the doc to Swift docc by @sebsto in #53
• Add docc on a github page by @sebsto in #54
• Fix error on doc action by @sebsto in #55
• Fix Doc GH Action by @sebsto in #56
• New Embedding Example by @sebsto in #57
• Add basic support for agent runtime by @sebsto in #58
• Bump glob from 10.4.5 to 10.5.0 in /Examples/web-playground/frontend by @dependabot[bot] in #59
• Add support for ServiceTier by @sebsto in #61
• Rework converse streaming files and add servicetier to streaming by @sebsto in #62

Full Changelog: 1.6.0...1.7.0