New version of Gemini beats other AIs at math, science, and reasoning
Google's new Gemini Pro is leading the way in artificial intelligence, surpassing its competitors in reasoning, science, and coding. The latest benchmark results released by Google on Thursday showcase the superior performance of Gemini 2.5 Pro in various key areas.
Benchmark Results
According to Google's findings, Gemini 2.5 Pro outperforms other top competitors such as OpenAI o3, Claude Opus 4, and others in key benchmarks like Humanity's Last Exam, Aider Polyglot, and FACTS Grounding.
Leadership Position
With a remarkable score of 1470, Gemini 2.5 Pro has secured the top spot on the LMArena leaderboard, underscoring its dominance in the AI landscape.
Availability
While the enhanced features of Gemini 2.5 Pro are generating buzz, the final version is not yet widely available. Google has labeled this release as an "upgraded preview," with a stable release expected to launch in the coming weeks. Interested users can access the preview version through the Gemini app.
