Sonnet 3.5 vs Llama 3.1 405B vs GPT-4omni: My TEST
Simple performance test what AI models have basic geographical knowledge for your holiday destinations. Claude 3.5 Sonnet, OpenAI GPT-4omni, GPT-4 Turbo, Mistral Large, Mixtral 8x22B, Llama 3 8B, Llama 3.1 405B, Gemini Advanced, Gemini Pro 1.5 Experimental 0801, ...
Trust AI for Your HOLIDAYS? Blind TEST provides first insights for basic geographical knowledge of beautiful and famous holiday destinations around Australia, New Zealand, Bali, Fiji, Hawaii, .....
Tested for zero shot knowledge and then with additional information (ICL) to help the AI find the correct answer. Live recording.
Test performed on LMsys.org (available for free to anyone):
https://arena.lmsys.org/
#airesearch
#tested
#newtechnology
