Anthropic has released Claude Opus 4.1, a new version of its advanced AI model. The upgrade brings better performance in real-world coding, reasoning, and research tasks. It is available now to all paid Claude users, including through Claude Code, API, Amazon Bedrock, and Google Cloud’s Vertex AI.
The company says Opus 4.1 shows major improvements over its previous version, Opus 4, especially in tasks that require tracking details and analyzing data. It also performs significantly better in complex coding tasks.
For example, Claude Opus 4.1 scored 74.5% on the SWE-bench Verified test, a key benchmark for AI coding ability. Companies like Rakuten and Windsurf say the new model is more accurate and helpful, especially for tasks like debugging and editing large code files without breaking them.
Despite the upgrade, pricing remains the same as Opus 4. Anthropic recommends that all users switch to Opus 4.1 to take full advantage of the improvements.
Developers can access the new model through the Claude API by using: claude-opus-4-1-20250805
.
Anthropic says bigger updates are coming soon and is encouraging user feedback to help shape future releases.