Google updates Android Bench with new LLMs, but Gemini still lags behind

Google has refreshed Android Bench, adding support for newer large language models and expanding its agent suite, including Fable‑5. The firm encourages developers to contribute to the benchmark’s evolution and provide data to improve its accuracy, while noting that Gemini continues to lag behind.

Summary written by Kernelia from the original article by Ars Technica — AI. The story and its rights belong to its author.