Loading...
Market
Which AI model will achieve highest accuracy in mathematics tasks by end of 2025?
2
Resolution / Starting Odds
Phi-4 • 25%
Gemini Pro • 25%
Llama 3.3 • 25%
Other • 25%
Results from mathematics-focused AI competitions or published research
Story
Microsoft Launches Phi-4 AI Model with 14 Billion Parameters, Competing with Gemini Pro and Matching Llama 3.3
Dec 13, 2024, 10:41 AM
Microsoft has introduced Phi-4, a new generative AI model featuring 14 billion parameters, which reportedly excels in advanced reasoning tasks, particularly in mathematics. Despite its smaller size compared to competitors like Google's Gemini Pro and Llama 3.3, Phi-4 demonstrates performance on par with these larger models, showcasing a significant advancement in AI efficiency. The model's effectiveness is attributed to its training on predominantly synthetic data, allowing it to achieve impressive results with fewer parameters. Microsoft Research's latest release positions Phi-4 as a contender in the evolving landscape of AI development, redefining expectations for model capabilities.
View original story
Similar markets
DeepSeek-R1-Lite-Preview • 25%
OpenAI's o1-preview • 25%
Google DeepMind's model • 25%
Other • 25%
OpenAI's model • 25%
Deepseek-Math (7B) • 25%
Gemini 1.5 Pro (May) • 25%
GPT-4o • 25%
GPT-5 • 25%
Claude 4 • 25%
Llama 4 • 25%
Other • 25%
OpenAI's model • 25%
Deepseek-Math (7B) • 25%
Gemini 1.5 Pro (May) • 25%
GPT-4o • 25%
Google's Gemini • 25%
OpenAI's GPT • 25%
Microsoft's Azure AI • 25%
Other • 25%
Meta's Llama 3.1-70B • 25%
OpenAI's GPT-4 • 25%
Google's Bard • 25%
Other • 25%
ChatGPT-4o • 25%
Google's Gemini • 25%
Another AI model • 25%
No clear leader • 25%
Llama 3.1 405B • 25%
GPT-4o • 25%
Claude Sonnet 3.5 • 25%
Other • 25%
Claude 3.5 Sonnet • 33%
GPT-4o • 33%
Google's AI Model • 33%
Claude 3.5 Sonnet • 33%
GPT-4o • 33%
Gemini • 34%
Yes • 50%
No • 50%
OpenAI o1-preview • 25%
Anthropic Claude 3.5 Sonnet • 25%
OpenAI o1 mini • 25%
Other • 25%