Gemini Pro preview - Even better coding performance

Developers had been doing amazing things with Gemini 2.5 Pro, so an updated version was released a couple of weeks early to get into developers’ hands sooner.

Gemini 2.5 Pro Preview (I/O edition) was recently released. This update features even stronger coding capabilities, for developers to start building with before Google I/O later this month. Meaningful improvements were made for front-end and UI development, alongside improvements in fundamental coding tasks such as transforming and editing code, and creating sophisticated agentic workflows.

Gemini Pro Preview: Even better coding performance

"We found Gemini 2.5 Pro to be the best frontier model when it comes to 'capability over latency' ratio. I look forward to rolling it out on Replit Agent whenever a latency-sensitive task needs to be accomplished with a high degree of reliability," said Michele Catasta, President, Replit.

improved coding performance on WebDev Arena

Best-in-class frontend web development

Gemini 2.5 Pro now ranks #1 on the WebDev Arena leaderboard, which measures human preference for a model’s ability to build aesthetically pleasing and functional web apps. Drawing on this leading capability, Gemini 2.5 Pro powers Cursor’s innovative code agent and empowers collaborations with companies like Cognition and Replit. Together, the frontiers of agentic programming are being pushed to unlock new possibilities for developers.

"The updated Gemini 2.5 Pro achieves leading performance on our junior-dev evals. It was the first-ever model that solved one of our evals involving a larger refactor of a request routing backend. It felt like a more senior developer because it was able to make correct judgement calls and choose good abstractions," said Silas Alberti, Founding Team, Cognition.

Gemini 2.5 Pro in action

Gemini 2.5 Pro’s deep understanding of code, combined with powerful reasoning, continues to make Gemini 2.5 Pro the go-to model for developers. The following cases are particularly exciting examples of how this model can be used.

Video to code

Gemini 2.5 Pro delivers state-of-the-art video understanding, scoring 84.8% on the VideoMME benchmark. Combining this with coding enables new flows that were previously not possible with earlier versions. For example, the Video to Learning App in Google AI Studio demonstrates how Gemini 2.5 Pro creates an interactive learning app based on a single YouTube video. With improved video understanding and complete UI, the updated Gemini 2.5 Pro model delivers a more functional experience than the previous simple example.

Easier feature development

Gemini 2.5 Pro is strong at front-end web development, helping developers get more done. Implementing new features means manually diving into design files and inspecting components to match style properties like colors, fonts, padding, margins, and borders then manually writing the CSS code needed to replicate those visual properties accurately. Now imagine using Gemini 2.5 Pro in an IDE and having the model generate new features, like adding a video player in the style of the other apps in the Gemini 95 starter app.

Quick concepts to working apps

Bringing ideas to life with both functionality and a beautiful UI is made easier with Gemini 2.5 Pro. The new dictation starter app, built using the updated model, is a great example of this in action. Some of the details such as the wavelength animations, responsive design, and subtle button hover effects stand out. By default, the model has a real taste for aesthetic web development while maintaining its steerability, helping developers quickly take a concept to a working web app. Gemini 2.5 Pro was able to design and code the microphone UI animation for the dictation starter app.

Start building with Gemini 2.5 Pro

Developers can build with Gemini 2.5 Pro using the Gemini API in Google AI Studio, and enterprise customers can use Vertex AI.

For developers already using Gemini 2.5 Pro, this new version not only improves coding performance but also addresses key developer feedback including reducing errors in function calling and improving function calling trigger rates. The previous iteration (03-25) now points to the most recent version (05-06), so no action is required to use the improved model, and it continues to be available at the same price. The model card has also been updated with the new version of 2.5 Pro.