Amazon Bedrock now offers Anthropic’s Claude Opus 4.7, a major upgrade aimed at production-grade workflows that demand stronger reasoning, longer context windows, and more reliable scaling. Opus 4.7 builds on the Opus family’s strengths and targets real-world use cases such as agentic coding, multi-step knowledge work, long-running tasks, and high-resolution visual understanding. Running on Bedrock’s next-generation inference engine, the model is positioned for enterprise deployments where performance, privacy, and availability matter.
What’s new in Claude Opus 4.7
Claude Opus 4.7 improves the model’s ability to reason through ambiguity, follow instructions precisely, and self-verify outputs. The release emphasizes better performance on long-horizon autonomy — useful for multi-step agents and complex systems engineering — while adding higher-resolution vision capabilities for charts, dense documents, and screen UIs. Opus 4.7 also supports a full 1M token context window, helping it stay coherent and on-task over long-running sessions.
Performance highlights
Anthropic reports measurable gains on several benchmarks and professional tasks:
- Agentic coding: Opus 4.7 continues to lead on agentic coding challenges, with scores such as 64.3% on SWE-bench Pro, 87.6% on SWE-bench Verified, and 69.4% on Terminal-Bench 2.0.
- Knowledge work: The model advances document creation, financial analysis, and multi-step research workflows, achieving 64.4% on Finance Agent v1.1.
- Long-running reliability: Improvements across the full token window make Opus 4.7 better at maintaining context and verifying outputs in extended interactions.
- Vision: Higher-resolution image understanding improves accuracy where fine detail matters.
Bedrock’s next-generation inference engine
Amazon Bedrock’s new inference engine underpins this launch with scheduling and scaling logic designed for predictable, production workloads. The engine dynamically allocates capacity to incoming requests, prioritizes steady-state workloads, and ramps resources quickly when demand shifts. During spikes, the system queues requests rather than rejecting them outright. Bedrock also provides “zero operator access,” ensuring customer prompts and responses remain inaccessible to Anthropic or AWS operators, helping protect sensitive data. Immediate per-account capacity is set at up to 10,000 requests per minute (RPM) per Region, with possibilities to request higher limits.
How to try Opus 4.7 in Bedrock
You can experiment with Claude Opus 4.7 directly in the Amazon Bedrock console by selecting Playground and choosing the model. Programmatic access is available through Anthropic’s Messages API (the Anthropic[bedrock] mantle client) or Bedrock-native APIs such as Converse and Invoke. Bedrock supports OpenAI-compatible Responses-style calls as well as lower-level Invoke operations. AWS provides Quickstart guidance and sample code in the console to generate short-term API keys for testing and to demonstrate calls via SDKs and the AWS CLI.
Practical tips and considerations
- Prompting and harnesses: Opus 4.7 is an upgrade over Opus 4.6 and may require prompt adjustments or harness tuning to unlock its best behavior in your applications.
- Adaptive thinking: Use Adaptive thinking when you need the model to allocate token budgets dynamically based on request complexity for more efficient and accurate reasoning.
- API choice: Choose the API that matches your needs — Converse for multi-turn workflows with guardrails, Invoke for fine-grained control, or the Anthropic Messages API for a streamlined Messages-based interface.
- Cost and scaling: Test steady-state and burst scenarios to understand performance and costs under your expected workloads; consult Bedrock pricing for region-specific details.
Availability and resources
Claude Opus 4.7 is available today in several AWS Regions, including US East (N. Virginia), Asia Pacific (Tokyo), and Europe (Ireland and Stockholm), with additional Regions announced over time. For more details and examples, consult the Claude by Anthropic in Amazon Bedrock page, Anthropic’s prompting guide, and the Bedrock pricing documentation. If you adopt Opus 4.7, provide feedback through AWS re:Post for Amazon Bedrock or your usual AWS Support channels.
Licensed under CC BY-ND 4.0 International
Anthropic’s Claude Mythos Leak: When Pre-Release Secrets Meet Cybersecurity Risk
Anthropic recently found itself at the center of an avoidable but consequential…
Google brings “Skills” to Chrome so Gemini prompts are instantly reusable
Chrome is getting another nudge toward becoming the home for Google’s AI…
Claude Opus 4.7: Anthropic’s Sharper, More Reliable Coding and Multimodal Model
Anthropic’s newest release, Claude Opus 4.7, is now generally available. Built as…
Two Words, One Deal: How “Stateful” vs “Stateless” Could Decide a $50 Billion Cloud Dispute
Last week’s reporting brought into sharp relief a narrowly technical — but…