Google's Gemini beats ChatGPT 4 on THESE benchmarks
Google recently launched Gemini, its latest and most powerful language model yet. It has outperformed ChatGPT in most benchmark tests. Google CEO, Sundar Pichai, expressed his excitement about the potential of AI.
OpenAI released GPT-4, its most powerful language model to date, earlier this year. It blew us all away by passing multiple benchmark tests. ChatGPT-4 seems to be good at everything else than literature. And just when we thought this was the pinnacle of AI, Google launched Gemini, its own LLM to compete with GPT-4.
Gemini was unveiled in May during Google's I/O event, and it piqued the interest of all tech fans. One of the things that piqued our interest was the LLM's multimodal capabilities, and we couldn't wait to put it to the test. That wait has now come to an end.
Google announced the debut of Gemini on Wednesday, calling it the "most capable and general model they have ever built." In a blog post, the business also stated that Gemini outperformed ChatGPT in most benchmark testing. Gemini was released in three variants: Ultra, Pro, and Nano.
Gemini was tested in the areas of General, Reasoning, Math, and Code to see how well it interacts with texts. Image, video, and audio benchmark tests were performed to assess the LLM's multimodal capabilities. And, with the exception of "HellaSwag," which is characterised as "common sense reasoning for everyday tasks," Gemini has outperformed ChatGPT in all tests.
The company also revealed that Gemini Ultra is the first model to "outperform human experts on MMLU (massive multitask language understanding), which uses a combination of 57 subjects such as math, physics, history, law, medicine, and ethics for testing both world knowledge and problem-solving abilities."
Google CEO, talking about Gemini in a blog post, said that we are probably seeing the most profound transition with AI and this is bigger than when we shifted to mobile phones from landlines or the internet. The Google CEO also said that AI can create more opportunities for people everywhere and drive innovation, economic progress, learning, creativity, and knowledge on a scale that we have not seen before.