Loading...
Market
Will the Strawberry AI model achieve a new benchmark score in MATH by January 31, 2025?
3
Resolution / Starting Odds
Yes • 50%
No • 50%
Official publications or announcements from OpenAI or benchmarking bodies
Story
OpenAI to Launch 'Strawberry' AI Model, Key to Orion/GPT-6, This Fall
Aug 27, 2024, 08:12 PM
OpenAI is preparing to launch a new AI model named 'Strawberry' this fall, aimed at enhancing its chatbot business. Previously known as Q*, the Strawberry model is designed to tackle complex math and programming problems, solve novel challenges, and produce synthetic data for training models. It has been demonstrated to U.S. national security officials and is expected to be integrated into ChatGPT. This model will also aid in the development of OpenAI's next-generation language model, 'Orion', also referred to as GPT-6. The Strawberry AI is set to surpass current capabilities, promising advanced reasoning and problem-solving skills, having already scored over 90% on the MATH benchmark.
View original story
Similar markets
Yes • 50%
No • 50%
Less than 90% • 25%
90%-94.99% • 25%
95%-99.99% • 25%
100% • 25%
Yes • 50%
No • 50%
Yes • 50%
No • 50%
Significantly better • 25%
Slightly better • 25%
On par • 25%
Worse • 25%
DeepSeek-R1-Lite-Preview • 25%
OpenAI's o1-preview • 25%
Google DeepMind's model • 25%
Other • 25%
Yes • 50%
No • 50%
Yes • 50%
No • 50%
Yes • 50%
No • 50%
OpenAI's model • 25%
Deepseek-Math (7B) • 25%
Gemini 1.5 Pro (May) • 25%
GPT-4o • 25%