Gemini 3.6 Flash crushes browser automation benchmarks at 68%

Google’s new Gemini 3.6 Flash is already making a strong first impression, especially if you’re using AI to control a web browser.

Browser Use, the company behind a popular browser automation framework, says Gemini 3.6 Flash reached a 68% task success rate on its BU Benchmark. That puts it ahead of GPT-5.6-sol, which scored 67%, and Claude Sonnet 4.6 at 62%. Only Claude Opus 4.8 ranked higher with 74%, though Browser Use says Opus comes at a much higher price.

“Gemini 3.6 Flash is here, and it’s the best price-to-performance model we’ve tested for browser agents,” Browser Use wrote in a post on X.

The benchmark measures how well AI models complete real browser tasks like clicking through websites, filling out forms, searching for information, and navigating pages without getting stuck. That’s becoming an increasingly important test as more companies build AI agents that can actually use the web instead of just answering questions.

If you’ve been following this space, Browser Use has been gaining attention for making browser automation much more affordable than traditional approaches. 

The results are also interesting because Gemini 3.6 Flash isn’t Google’s flagship model. It’s designed to balance speed, quality, and cost. According to Google DeepMind, the new model uses fewer tokens than Gemini 3.5 Flash while delivering better results at the same price. So this makes it a practical option, especially when you consider that every click, page load, and reasoning step adds to the final bill.

Browser Use also shared results for the other newly announced models. Gemini 3.5 Flash scored 58%, while Gemini 3.5 Flash-Lite came in at 49%. GPT-5.4-mini finished last in the chart with 36%.

For now, though, Gemini 3.6 Flash looks like the new value leader. It doesn’t top the chart outright, but getting within six percentage points of Claude Opus 4.8 while costing far less is likely to catch the attention of developers building AI agents that spend most of their time inside a browser.

That said, at the time of this writing, Gemini 3.6 Flash hasn’t been added to Browser Use Cloud.

We stand out from the tech-media crowd because we break news stories; we mainly bring you stuff that you won’t find anywhere in the mainstream tech media. Our stories have been picked up by some of the world’s most popular websites and media outlets—more info is available here.