Summary:
Crowd-trained approach to LLMs could disrupt the AI industry.
New model Collective-1 created by startups Flower AI and Vana.
Flower AI enables training on distributed computers over the internet.
Future models planned with 30 billion and 100 billion parameters.
Vana allows users to share personal data for training, enhancing data sources.
A New Era in AI Development
A groundbreaking crowd-trained approach to developing large language models (LLMs) over the internet is set to disrupt the AI industry. This innovative method could lead to the release of a 100 billion-parameter model later this year.
The Collective-1 Model
Startups Flower AI and Vana have collaborated to create a new model known as Collective-1, which leverages distributed computing to train AI. This model utilizes private and public data, indicating a shift in how AI is developed.
Flower AI’s technology enables training across numerous computers connected via the internet, allowing companies to develop AI models without needing centralized data or compute resources. Vana contributes data from platforms like X, Reddit, and Telegram to enhance training.
Scale and Future Plans
While Collective-1 currently has 7 billion parameters, Flower AI is working on a model with 30 billion parameters and aims for a 100 billion-parameter model soon. This distributed approach has the potential to reshape the AI landscape, making it feasible for smaller companies and universities to participate in advanced AI development.
Rethinking AI Training
Traditional AI models are built using massive amounts of data housed in data centers with advanced GPUs. The new distributed method allows computations to occur across various locations, potentially transforming the power dynamics in the AI field.
The Role of User Data
Vana is also pioneering a method for users to share their personal data for AI training, granting them control over its usage and potential financial benefits. This could unlock new data sources, especially in sensitive areas like healthcare and finance.
Conclusion
Flower AI’s innovative methods promise to democratize AI development, enabling broader participation and potentially leading to new, powerful models that leverage decentralized data. This shift could significantly impact how the AI industry evolves in the coming years.
Comments