%20--%3e%3cdefs%3e%3cstyle%3e%20.st0%20{%20fill:%20%23061b40;%20}%20.st1%20{%20fill:%20%23306af1;%20}%20.st2%20{%20fill:%20%235ce5cf;%20}%20%3c/style%3e%3c/defs%3e%3cg%3e%3cpath%20class='st0'%20d='M55,10.5h9v3h-9v9h12V7.5h-12v3ZM64,19.5h-6v-3h6v3Z'/%3e%3cpolygon%20class='st0'%20points='69%2016.5%2078%2016.5%2078%2019.5%2069%2019.5%2069%2022.5%2081%2022.5%2081%2013.5%2072%2013.5%2072%2010.5%2081%2010.5%2081%207.5%2069%207.5%2069%2016.5'/%3e%3cpolygon%20class='st0'%20points='95%2010.5%2095%207.5%2083%207.5%2083%2022.5%2095%2022.5%2095%2019.5%2086%2019.5%2086%2016.5%2095%2016.5%2095%2013.5%2086%2013.5%2086%2010.5%2095%2010.5'/%3e%3cpath%20class='st0'%20d='M40,1.5v21h11.6l1.4-1.4v-7.6h0c0,0-1.4-1.5-1.4-1.5l1.4-1.4V2.9l-1.4-1.4h-11.6ZM50,19.5h-7v-6h7v6ZM50,10.5h-7v-6h7v6Z'/%3e%3c/g%3e%3cpath%20class='st1'%20d='M23.1,24L14.7,4.8l-4.9,11.2h3.8l-1.8,4H3.3L12.1,0H2C.9,0,0,.9,0,2v20c0,1.1.9,2,2,2h21.1Z'/%3e%3cpath%20class='st2'%20d='M34,0h-16.8l10.6,24h6.2c1.1,0,2-.9,2-2V2C36,.9,35.1,0,34,0ZM32.5,20h-4V4h4v16Z'/%3e%3c/svg%3e)
Microsoft GitHub announced an exciting update on April 6th, introducing an experimental feature called Rubber Duck for its Copilot CLI. This new feature introduces a "second opinion" review mechanism across models, aimed at helping developers improve the accuracy and efficiency of their code, ultimately boosting AI performance by nearly 75%.
During software development, early decision errors often accumulate into larger issues, and traditional self-review methods are prone to be influenced by the model's own training biases. The introduction of Rubber Duck aims to bring in different models as independent reviewers, offering diverse perspectives and identifying potential errors promptly. This feature allows users to select the Claude series model as the main controller, followed by GPT-5.4 for code review, ensuring the accuracy and completeness of the code.
According to research using the SWE-Bench Pro benchmark, there was a significant performance gap between the standalone operation of Claude Sonnet 4.6 and Opus 4.6. However, after combining Sonnet 4.6 with Rubber Duck, it successfully closed 74.7% of the performance gap. The feature performs even more notably when handling complex tasks, scoring 3.8% higher than the baseline. Specific cases show that it can effectively identify architectural logic vulnerabilities, loop coverage errors, and conflicts across files.
Rubber Duck offers flexible usage options, supporting three types of review modes: active, passive, and user-triggered. The system automatically seeks reviews at key points such as planning, complex implementation, and test writing, and can also be triggered passively when developers face problems. Additionally, users can request reviews at any time, and the system will display feedback content and modification justifications.
Currently, the Rubber Duck feature is available in experimental mode. Users can simply install GitHub Copilot CLI and run the /experimental command to activate this new feature and experience the collaborative work of the Claude model and GPT-5.4.
Key Points:
🌟 1. Microsoft GitHub introduced the Rubber Duck feature, implementing a cross-model AI review mechanism to enhance programming efficiency.
🔍 2. This feature successfully closed 74.7% of the performance gap by combining Claude Sonnet 4.6 and GPT-5.4.
⚙️ 3. Rubber Duck supports multiple review modes, allowing users to proactively request reviews to ensure code quality.