Google is teaching Gemini how to handle a lot of code at once

Google is planning to give Gemini a major boost in absorbing and explaining software code, as discovered in as-yet unreleased code by Android Authority. The upgrade could simplify developers’ lives using the AI assistant on Android devices.

Specifically, the unreleased programming points to Gemini, adding the ability to analyze entire code folders instead of one file at a time like now. That’s less of a breakthrough than a way for Google to keep up with its rivals like ChatGPT, which can already take on multiple files of code.

Based on the report, Gemini users would be able to upload a codebase of up to 1,000 files and 100MB in a single folder. You could then ask Gemini about the code, what it does, and ways to improve it. That might not seem like a big change, but the single file limit makes Gemini’s insight into code much less helpful for developers.

Even if they like how Gemini explains the code, uploading a thousand files in a folder that could all go up at once takes a long time. By putting it all together in a single upload, Gemini could provide a more holistic view of a project and its strengths and weaknesses. At the very least, debugging would go a lot faster.

The raw code doesn’t indicate when the new feature will be available or if it will be restricted to premium subscribers, but expect a lot of excitement from coders who would prefer to use Google’s AI over OpenAI or GitHub Copilot, the coding assistant run by the Microsoft subsidiary.

Code Gemini

Even with folder uploading as an option, Gemini won’t be taking over for programmers everywhere. A maximum of 1,000 files and 100MB per folder might be sufficient for smaller projects or individual developers, but it could fall short for larger, enterprise-level codebases.

Even on a smaller scale, AI tools can really only supplement human expertise. Debugging and optimizing code is one thing, but you’ll still want a human checking the result of anything the AI writes. And the creative aspect of programming is not an area where AI can really shine, at least without mimicking existing ideas.

Still, many independent developers and small teams will undoubtedly welcome Gemini’s upgrade as it makes coding more accessible to those with less experience and fewer resources. Imagine a junior developer using Gemini to understand a complex open-source project or a solo creator resolving all of the bugs in their app. The ability to ask plain-language questions about an entire codebase could streamline that process enormously.

You might also like

• Gemini will yada yada your Google Chat into a neat summary
• Google is making Gemini AI part of everything you do with your smartphone – here’s how
• How to build your own customized Google Gemini AI chatbot

Eric Hal Schwartz is a freelance writer for TechRadar with more than 15 years of experience covering the intersection of the world and technology. For the last five years, he served as head writer for Voicebot.ai and was on the leading edge of reporting on generative AI and large language models. He’s since become an expert on the products of generative AI models, such as OpenAI’s ChatGPT, Anthropic’s Claude, Google Gemini, and every other synthetic media tool. His experience runs the gamut of media, including print, digital, broadcast, and live events. Now, he’s continuing to tell the stories people want and need to hear about the rapidly evolving AI space and its impact on their lives. Eric is based in New York City.