-
-
Catalogue of LLMs to try out
-
Earnings and Usage Charts
-
Settings page to buy credits or get API Keys
-
Register new LLM endpoints to start earning
Inspiration
The inspiration behind dInference stems from the recognition of the significant computational resources required for Large Language Model (LLM) inferencing. Traditional centralized approaches often limit accessibility and concentrate power in the hands of a few. By leveraging decentralized technologies, the project seeks to democratize access to LLM inferencing capabilities, empowering a broader community of individuals and organizations.
What it does
dInference establishes a decentralized infrastructure for LLM inferencing, enabling GPU providers to contribute their resources and earn rewards in the form of dInference tokens. Through a web application, end-users can access LLM inferencing APIs distributed across the decentralized network, promoting accessibility and fostering innovation in the web3 ecosystem.
How we built it
We built dInference using a combination of web development technologies, Docker for containerization, smart contracts for token incentives, and load balancing techniques for efficient API distribution. The platform was meticulously designed and developed to ensure seamless integration of GPU resources, user-friendly experience, and robust performance.
Challenges we ran into
One of the primary challenges we encountered was optimizing the load balancing and API gateway components to handle varying workloads efficiently while maintaining high availability and low latency. Additionally, ensuring the security and integrity of the decentralized network posed challenges that required thorough consideration and implementation of robust protocols.
Accomplishments that we're proud of
We're proud to have successfully launched the dInference platform, providing a decentralized solution for LLM inferencing that promotes accessibility, decentralization, and innovation. The development of a self-contained Docker image and the implementation of a token economy to incentivize GPU providers represent significant accomplishments in our journey.
What's next for DInference | A decentralised LLM inferencing infrastructure
Looking ahead, we plan to further optimize and expand the dInference ecosystem through strategic partnerships, collaborations, and community engagement. We aim to enhance the platform's capabilities, improve performance, and explore new avenues for innovation in LLM inferencing and beyond within the web3 landscape.
Log in or sign up for Devpost to join the conversation.