Amazon Bedrock Launches Console for 3-Way Model Comparisons Across Anthropic and OpenAI APIs
Updated
Updated · AWS Blog · Jun 5
Amazon Bedrock Launches Console for 3-Way Model Comparisons Across Anthropic and OpenAI APIs
3 articles · Updated · AWS Blog · Jun 5
Summary
Amazon Bedrock rolled out a new console experience for its bedrock-mantle endpoint, letting developers test, compare and deploy the latest GPT, Claude and open-weight models through Anthropic Messages, OpenAI Responses and Chat Completions APIs.
Up to 3 models can be compared side by side, while new model cards surface capabilities, context windows, quotas, pricing and regional availability in one catalog aimed at speeding model selection.
Project-based dashboards track inference requests, errors and token usage, and live API documentation auto-fills model ID, Region, endpoint URL and API key references so code snippets can run without modification.
The console also supports connecting coding agents including Claude Code, Cline, Codex, Cursor and OpenCode, routing their requests through Bedrock with IAM credentials or Bedrock API keys.
The new experience is now available in all Regions offering bedrock-mantle, including US East, US West, parts of Asia Pacific and Europe, and São Paulo in South America.