New Claude 3.5 DESTROYS OpenAI's GPT-4o In All Benchmarks!

Claude 3.5 Sonnet, the latest AI model from Anthropic, delivers outstanding performance on benchmark tests, outperforming even GPT-4 in several domains. This sophisticated AI model demonstrates exceptional abilities in complex coding, humor understanding, and chart analysis, making it an invaluable resource across diverse applications. With competitive pricing and an intuitive user interface, Claude 3.5 Sonnet is quickly establishing itself as a formidable player in the fast-paced AI industry.
#ai #claude

[@nonunga]

i have used sonnet 3.5 mostly for front end development and js, its so much better that gpt4o. Also its vision capabilities are superior from my testing. A down side would be that you can reach its usage limits fairly quickly compared to gpt4o, even in the paid version. still worth it.

[@thymenwestrum7011]

I tried this question which both gpt and sonnet 3.5 get wrong intrestingly they both choose milan: Of the "Big Four cities that host one of the fashion weeks, which one is the farthest south? The correct answer is New York

[@nonunga]

Try using a better prompt, adding more context perhaps. Also these models are trained with data up to a certain date. For questions of that style, perplexity might be better.

[@thymenwestrum7011]

@@nonunga That's false. The system can accurately retrieve all the information it needs but struggles to determine the southernmost city. Even though it gets the latitude and longitude coordinates right.

[@miramarkovic9551]

Please include a translation for Croatian, Serbian or Bosnian. Please, I want to follow what you are saying.🙏
Thanks in advance.

[@TechRevolution-dj1zg]

We will do it in future videos.
Thanks for your Support and love

[@miramarkovic9551]

@@TechRevolution-dj1zg Well done. Thank you❤️