← Back
AWS ML Blog

Access OpenAI GPT models on Amazon Bedrock from Canada with cross-Region inference

•15 min read•
#bedrock#llm#deployment
Level:Advanced
For:AI Engineers
✦TL;DR

Amazon Bedrock now enables Canadian customers to invoke OpenAI GPT‑5.6, GPT‑6, and GPT‑6.1 models from Canada (Central) and Canada West (Calgary) via cross‑Region inference, routing requests to US regions while preserving North‑America data residency. The models (e.g., GPT‑6.1 Sol, GPT‑6 Astra, GPT‑5.6 Luna) support text + image inputs, a 1‑million‑token context window, reasoning mode, tool calling, and prompt caching. Engineers can call these models through the OpenAI Responses API, Chat Completions API, or Amazon Bedrock Converse API on the `bedrock-runtime` endpoint without managing destination routing. This expands capacity and helps sustain throughput during demand spikes for production generative‑AI workloads.

⚡ Key Takeaways

  • Models available: GPT‑6.1 Sol, GPT‑6 Astra, GPT‑6 Sol, GPT‑6 Luna, GPT‑5.6 Sol, GPT‑5.6 Terra, GPT‑5.6 Luna.
  • Inference profiles: `us.*` (US geographic) and `global.*` logical identifiers replace raw model IDs.
  • Prompt caching is supported for GPT‑5.6, GPT‑6, and GPT‑6.1 on Bedrock.
  • Calls are made to the `bedrock-runtime` endpoint using OpenAI Responses API, Chat Completions API, or Bedrock Converse API.
  • Data residency: processing occurs in North America; cross‑Region routing is handled by Bedrock, not the application.
💡 Why It Matters

Engineers can now leverage the latest OpenAI foundation models from Canadian regions without redesigning their routing logic, gaining access to larger capacity pools and maintaining compliance with North‑American data‑residency requirements.

✅ Practical Steps

  1. Configure your IAM permissions to allow `bedrock-runtime:InvokeModel` in the Canadian source Region.
  2. Invoke a model by calling the OpenAI Responses API, Chat Completions API, or Bedrock Converse API on the `bedrock-runtime` endpoint, specifying the appropriate inference profile (e.g., `us.gpt-6.1-sol`).
  3. Request a quota increase via the AWS console if you anticipate demand spikes, then monitor observability signals in your source Region.

Want the full story? Read the original article.

Read on AWS ML Blog ↗

More like this

Censored vs. Truncated Regression in Python, Explained

Towards Data Science•#llm

New in AWS Quick: More capable flows, an updated web experience, and agents that work together

AWS ML Blog•#agents

Three insights you may have missed from theCUBE’s coverage of the Dell Technologies New York Symposium

SiliconANGLE AI•#deployment

How Google Duplex evolved to also call businesses for you in Gemini

Google Research•#agents

EXPLORE AI NEWS

Daily hand-picked stories on LLMs, RAG, agents and production AI — curated for engineers who ship.

BROWSE NEWS

GET THE WEEKLY DIGEST

Join engineers getting the Monday signal-over-noise AI breakdown. No spam, unsubscribe anytime.

LEARN AI ENGINEERING

Curated courses, research papers, repos and tutorials built for engineers leveling up in AI.

START LEARNING