Google Announces Gemini 2.5 Pro (I/O Edition) With Coding Upgrades

Google is updating Gemini 2.5 Pro with “significantly improved” coding capabilities. This updated model was originally going to be released at I/O 2025, but the company decided to move up the launch.

Cheekily referred to as “Gemini 2.5 Pro (I/O Edition),” Google touts improvements in “code transformation, code editing, and developing complex agentic workflows.” It follows the initial (03-25) release in late March.

Google is particularly touting how Gemini 2.5 Pro surpassed the previous #1 on the WebDev Arena Leaderboard by 147 Elo points. (2.5 Pro was previously at #2.) In terms of benchmarks, LiveCodeBench v5 (code generation) goes from 70.4% to 75.6%.

This leaderboard measures human preference for a model’s ability to build aesthetically pleasing and functional web apps.

Google also touts ”state-of-the-art performance in video understanding, with a score of 84.8% on the VideoMME benchmark.” Developers will also benefit from reduced errors when function calling and improved 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. We have also updated the model card with the new version of 2.5 Pro .

This update to 2.5 Pro (gemini-2.5-pro-preview-05-06) is available in the Gemini app, with Canvas particularly benefiting (as seen below), and the Gemini API (Google AI Studio + Vertex AI). Google moved up the release by two weeks given “overwhelming enthusiasm for this model.”

Google teases ‘exciting’ Gemini updates at I/O 2025, like ‘more personalized assistant’
Google wants Gemini to be your ‘personal, proactive, and powerful’ assistant