Which capability will Google's Gemini-exp-1206 lead in Chatbot Arena by end of 2025?
Coding • 25%
Math • 25%
Instruction Following • 25%
Style Control • 25%
Chatbot Arena leaderboard rankings
Google's Gemini-Exp-1206 Model Dominates Chatbot Arena with 2M Token Context Window
Dec 6, 2024, 05:10 PM
Google DeepMind has released a new experimental model, Gemini-exp-1206, which has quickly risen to the top of the Chatbot Arena leaderboard, surpassing competitors like OpenAI's models. This model, available for free in Google AI Studio and the Gemini API, features a 2 million token context window, marking a significant advancement in AI capabilities. The release coincides with the one-year anniversary of Google's first Gemini model, highlighting the company's ongoing commitment to AI innovation. The new model has achieved first place in overall ranking, as well as in categories such as hard prompts, coding, math, instruction following, and style control. This development intensifies the competition in the AI sector, with Google positioning itself as a leader in AI technology.
View original story
No • 50%
Yes • 50%
OpenAI • 25%
Microsoft AI • 25%
Other • 25%
Meta AI • 25%
OpenAI • 25%
Anthropic • 25%
Other • 25%
Meta • 25%
400 million to 500 million • 25%
Less than 300 million • 25%
300 million to 400 million • 25%
More than 500 million • 25%
Other • 25%
Google • 25%
OpenAI • 25%
Microsoft • 25%
Microsoft • 25%
OpenAI • 25%
Meta • 25%
Other • 25%
Apple introduces similar feature • 25%
Amazon introduces similar feature • 25%
No major competitor response • 25%
Samsung introduces similar feature • 25%
OpenAI's o1 • 25%
Other • 25%
Microsoft's AI Model • 25%
Gemini 2.0 Flash Thinking • 25%
No • 50%
Yes • 50%
No • 50%
Yes • 50%
Microsoft • 25%
Google • 25%
Other • 25%
OpenAI • 25%