Fierce competition continues in the world of artificial intelligence. OpenAI GPT-4 Although the based model was shown as the most powerful model of recent times, new tests revealed that it was left behind. Alright best artificial intelligence Which model?
Claude-3 vs ChatGPT-4: Best AI model
LMSys Chatbot Arena Leaderboard, which tracks the performance of artificial intelligence systems, is developed by Anthropic. Claude-3 Opus revealed that the model received the highest score. This shows that GPT-4, which has been the leader for many years, is left behind.
The Leaderboard was judged by judges during “chat battles”. The judges tested the chatbots by entering various commands from different topics and used the Elo rating system for comparison.
New era in Tesla cars! Full charge in 15 minutes
Tesla and CATL are working on a new battery. It is said that the batteries developed by CATL reach full charge in just 15 minutes.
Claude-3 Opus came out on top with 1253 Elo points in the latest ratings update. The second and third places were OpenAI GPT-4 preview models with 1251 and 1248 points, respectively.
In the rest of the rankings, the Google Gemini Pro model came in fourth place with 1203 points, and Anthropic’s previous Claude-3 Sonnet model came in fifth place with 1198 points. In addition, the Qwen1.5 model of the Chinese company Alibaba ranked ninth with 1148.
Here is the list of the best artificial intelligence models:
- Claude 3 Opus – 1253
- GPT-4-1106-preview – 1251
- GPT-4-0125-preview – 1248
- Bard (Gemini Pro) – 1203
- Claude 3 Sonnet – 1196
- GPT-4-0314 – 1193
- Claude 3 Haiku – 1181
- GPT-4-0613 – 1174
- Mistral-Large-2402 – 1153
- Qwen1.5-72B-Chat – 1153
GPT-4 set a very high standard when it was released last year. However, Anthropic developers seem to have managed to surpass GPT-4 in their model training. Claude Opus also broke a record by scoring +5/-5 in the chat battle.