OpenAI GPT-6.1 Terra arrives on Amazon Bedrock with lower token pricing

OpenAI GPT-6.1 Terra Arrives on Amazon Bedrock With Reduced Token Pricing

Amazon Web Services released OpenAI’s GPT-6.1 Terra for general availability on Amazon Bedrock, featuring input and output token pricing set at least 50% lower than both GPT-6.1 Sol and GPT-5.6 Terra, according to reporting on daily.dev. Terra positions itself as the everyday-workload line within the GPT-6 family, handling up to 1 million input tokens across text and image formats through Bedrock APIs, ChatGPT Work, and Codex.

The Tech TL;DR:

GPT-6.1 Terra launches on Amazon Bedrock at half the token cost of previous models while handling up to 1 million input tokens.
AWS Step Functions connect directly to AI agents via AgentCore Gateway and the Model Context Protocol without requiring a rewrite of existing logic.
Containerized workloads and security architectures receive updates including ML-KEM post-quantum encryption, GPU sharing via NVIDIA MPS, and SOCI lazy loading.

Step Functions Integration Expands Agent Capabilities via Model Context Protocol

AWS integrated Step Functions workflows as callable tools for AI agents through the Amazon Bedrock AgentCore Gateway, which communicates over the Model Context Protocol (MCP). Both Standard and Express workflows qualify as connector targets without code rewrites. The architecture preserves strict operational sequencing: the agent determines when to execute a workflow and supplies the input parameters, while Step Functions handles individual execution steps sequentially. Support agents can trigger refunds while eligibility checks and manager approvals occur in their predefined order.

[AWS Update #220] OpenAI GPT-6.1 Terra Now Generally Available on Amazon Bedrock and More

Engineering teams connecting microservices to LLMs face significant token overhead when publishing large API catalogs. MoEngage bypassed this limitation by connecting roughly 400 internal REST API operations across seven Regions to AgentCore Gateway. Rather than transmitting a 20,000-token tool listing per connection, agents query tools by describing specific tasks in plain text, saving an estimated six months of platform engineering.

AWS IAM Identity Center Expands Identity Propagation and Adds Kubernetes Gateway API Support

apiVersion: gateway.networking.k8s.io/v1
kind: HTTPRoute
metadata:
  name: app-route
  namespace: production
spec:
  parentRefs:
  - name: internal-gateway
  rules:
  - matches:
    - path:
        type: Prefix
        value: /api
    backendRefs:
    - name: core-service
      port: 8080

Always consult with certified IT and cybersecurity professionals before altering enterprise networks or handling sensitive data.