Google updates best AI models for coding Android apps, Gemini & GPT 5.4 at the top

Ben Schoon | Apr 9 2026 - 10:05 am PT

The “Android Bench” for ranking AI models used in Android app development has been updated, with OpenAI’s latest model now tied with Gemini for the top spot.

First released in March, the “Android Bench” is Google’s resource for measuring the best AI models to use for coding Android apps. Google’s methodology includes looking at how the models can work with Jetpack Compose for UI, Coroutines and Flows for asynchronous programming, room for persistence, and hilt for dependency injection, among other factors.

In the first update to this list, Google has added two new models in OpenAI’s GPT 5.4 and GPT 5.3 Codex, and they quickly jump towards the top of the list.

Best AI for Android app development, according to Google (4/9/26)

New: GPT 5.4: 72.4%
Gemini 3.1 Pro Preview: 72.4%
New: GPT 5.3-Codex: 67.7%
Claude Opus 4.6: 66.6%
GPT-5.2 Codex: 62.5%
Claude Opus 4.5: 61.9%
Gemini 3 Pro Preview: 60.4%
Claude Sonnet 4.6: 58.4%
Claude Sonnet 4.5: 54.2%
Gemini 3 Flash Preview: 42%
Gemini 2.5 Flash: 16.1%

The rest of the list didn’t change this time around, with the results used still from late February in that initial run. OpenAI’s latest models were tested in mid-March ahead of this week’s release of those results.

Advertisement - scroll for more content

Of course, these results shouldn’t be treated as an absolute fact. As with any benchmark, reality often differs from controlled tests. There are a ton of variables for why one model might work better for you than another, including workflow, value, and more.

Google originally said that its goal in publishing these results was to help developers be “more productive” and, ultimately, deliver “higher quality apps across the Android ecosystem.”