ChatGPT is no longer the “king” in the AI world

ChatGPT: New version of Claude offers processing speed and intelligence beyond GPT-4. However, the applicability of this tool is not much.

Since its release, the title of first generative AI has been held by GPT-4. This tool ranks #1 in the Important Reviews category. In particular, OpenAI’s products are far ahead of competitors in terms of “vibes” (roughly translated as: emotions). From the moment GPT-4 was launched, most users who have spent time using and comparing major language models have agreed on GPT-4’s superior performance.

However, recent major updates from other LLMs have shaken ChatGPT’s dominance. 4 new products launched in one week, with benchmark scores approaching and exceeding OpenAI’s tools. Especially their relationship is also very good.

Among them, experts spoke highly of the anthropic version of Claude No. 3’s work. The solution’s ability to express emotions is very powerful. It is believed to be able to surpass GPT-4 by a significant margin. in LL.M. In benchmark tests, the Claude 3 Opus achieved leading scores in every category. It shows excellent reasoning skills, mathematical solving skills…

AI expert Simon Willison said he has switched to Claude 3 for most tasks, replacing OpenAI’s product. He especially appreciates the ability to use code in programming.

“I recently had issues typing complex commands that caused GPT-4 to generate JavaScript code that failed. However, when I ran it with Claude 3, I got the perfect answer,” said Simon Willison.

TechCrunch praises Anthropic for its document summarization solution. As a result, AI models often suffer from disproportionate input errors. Claude 3 achieved the best results among the LLM versions tested.

“Claude 3 Opus is one of the most effective chatbots I’ve ever used. At least the answers it gives are concise, easy to understand, and don’t contain specialized words,” commented TechCrunch reporter Kyle Wiggers.

However, Anthropic’s tools do not extend to handling website data or other inputs. It only answers questions about issues that occurred before August 2023. Meanwhile, Claude 3 Opus also lags behind other chatbots when it comes to recent events.

GPT-4 can handle many types of input data. Gemini 1.5 can analyze videos. Open Google’s tool to access your Gmail mailbox. These are things Anthropic hasn’t developed for Claude 3 yet. While it offers faster, more accurate, and smarter processing, the product’s $20/month price is on par with GPT-4 and Gemini. .

Viết một bình luận