This document serves as a supplement to the initiatives outlined in Phase 2 of the AI Video Subnet endeavor (refer to the pre-proposal). While not comprehensive, it delineates our primary focus areas. Kindly be aware that this list remains subject to potential adjustments influenced by evolving development circumstances or minor shifts in priorities.

AI SPE Roadmap Timeline.jpg

Additional Resources

livepeer-ai-spe changelog | Productlane

livepeer-ai-spe roadmap | Productlane

Back-end Requirements

In Phase 2 (Network Optimization Phase), our goal on the back-end side is to significantly enhance our AI Subnet core software and pipelines across multiple dimensions:

  1. Enhance AI Subnet Performance and Service Quality: Our objective is to substantially improve the efficiency, speed, and quality of outputs within the AI subnet.

  2. Clear Entry Point for App Developers: Provide a well-defined access point for App developers who are looking to start building on the network.

  3. Expand GPU Compatibility: Enable GPUs with lower VRAM and server GPUs to connect to our subnet, broadening our hardware support and increasing network capacity.

  4. Refine Pricing Structure and Enhance Transparency: We aim to optimize how we structure pricing and communicate it to users, ensuring both clarity and fairness.

  5. Provide Clear Metrics: Provide transparent AI Subnet job metrics, encompassing model usage and latency. This will empower orchestrators to select supported models wisely and assist Apps in choosing the most suitable models for their needs.

  6. Expand Model and Pipeline Offerings: We plan to introduce new models and develop additional pipelines, thereby enhancing our service offerings and accommodating a wider range of applications.

  7. Foster Open-Source Collaboration: Establish clear and accessible pathways for open-source contributors to engage with and contribute to the AI subnet development effectively.

  8. Enhance AI Subnet Performance and Service Quality

During Phase 2, our primary focus is enhancing the supply-side capabilities of the network. This involves reassessing and refining the existing implementations utilized in our Phase 1 (network Stability Phase) proof of concept (POC) to ensure they are optimized.