GitHub Copilot, the subject of Claude's 3.5 Sonnet on GitHub Copilot
GitHub has announced the release of Claude 3.5 Sonnet on GitHub Copilot, a coding tool designed to assist developers in their tasks. The new version is a significant upgrade, outperforming all publicly available models on SWE-bench Verified and achieving the top score in its class (93.7%) on HumanEval.
Claude 3.5 Sonnet is powered by Amazon Bedrock's cross-region inference, enhancing its reliability. This integration aims to target over 100 million developers on GitHub, making coding assistance more accessible than ever.
The tool, developed by Anthropic, is capable of writing correct Python functions from natural language descriptions. It can be found in the GitHub blog and documentation, with the public preview available starting today.
By the coming weeks, Claude 3.5 Sonnet will be accessible to all developers and organizations using GitHub Copilot Chat. It's important to note that while the specific details about how to access it are not provided, the tool is available for coding in Visual Studio Code and GitHub.com.
Claude 3.5 Sonnet is an upgrade from a previous version and serves as the base coding agent used in GitHub Copilot. Anthropic, the company behind Claude Sonnet models, has released Claude Sonnet 4, improving code quality, but Claude 3.5 Sonnet continues to be a valuable asset in previous versions of Copilot.
For more information about Claude 3.5 Sonnet, its capabilities, and how to use it, refer to the GitHub blog and documentation. Start exploring this powerful tool today and enhance your coding experience on GitHub Copilot.
