🍃 Lilypad Project Report: Jan 22
Growing the Lilypad open compute network, more modules, the Open D/I Summit, and more!
Table of contents
- 🌐 Overview
- ⚒️ Lilypad Engineering Update
- 🌟 Lilypad "All the Things" Update
- 🔮 What's Next?
- ☎️ Contact Us
Prompt: bright rolling hills with a futuristic city landscape in the background
Image made with Lilypad AI Studio :)
Our team is proud to help accelerate the democratization of AI and AI compute with the Lilypad serverless distributed compute network. The network provides infrastructure for use cases like AI inference, ML training, DeSci and more.
The Lilypad Calibration test-net continues to grow making more GPUs available to the community! We are super grateful for the early Calibration test-net contributors and continue to push towards the launch of the Lilypad incentivized test-net. Sign up to participate in the Calibration test-net here!
⚒️ Lilypad Engineering Update
Voice sample/text to voice module on Lilypad!
The team can't help but highlight our awesome community contributors! Fabri from Dub3 contributed a Voice Sample & Text to Voice module running on the Lilypad network. This module takes a voice sample as an input and can then output another voice sample for a given text prompt - in multiple languages.
Automating the generation of specs - bacalhau job on Lilypad network.
Image name and the corresponding container build version.
And the values of the pre-populated input entry points corresponding to the given container that you want to execute.
It then generates the corresponding spec module which can then be committed to the github repo and then executed as shown in the docs. This can reduce major barriers for non-developers and specialists that want to run the compute jobs on top of Lilypad avoiding:
Memoize the syntax of specs.
The pain of doing type checking due to the deserialization bugs in the specification when the build breaks due to updating the container of similar tags.
This tooling helps to usher the devops adoption of the bacalhau cloud (the traditional cloud has the creation of packages that were able to generate and validate yaml descriptions based on the infrastructure specifications of deployment).
🌟 Lilypad "All the Things" Update
The Open D/I Summit 2023
Ally Haire from Lilypad Network gave a keynote at the Open D/I Summit! The presentation highlights Lilypad's network architecture, core values, as well as the roadmap/vision. The talk begins at 3:00:06/3:14:55 in the video above.
Lilypad at LabWeek 2023 - Unlocking Powerful Distributed Compute for AI, ML & Web
Our team had a blast in Istanbul for Lab Week 2023 and ETH Devconnect! Check out a presentation from the team discussing "Lilypad - accessible hardware via an open, P2P marketplace with efficient pricing and distribution." Learn about executing AI jobs from Smart Contracts as well as the value of distributed compute for AI!
The presentation slide can be found here.
Dive into Lilypad!
Learn more about the Lilypad Network with technical demos, read about the vision, and get into use cases in the docs! To view currently supported modules, run a node, or create a module read more here.
The Lilypad Aurora v2 testnet supports modules including Mistral LLM, Stable Diffusion XL (SDXL 0.9), LoRA fine tuning, Filecoin Data prep, and more.
🔮 What's Next?
We continue to stress test the Lilypad Calibration test-net with different workloads and a variety of GPUs. The team is collaborating with the community adding more modules to run on the network.
☎️ Contact Us
Follow along on X/Twitter for the latest news and updates!
Did you find this article valuable?
Support Lilypad Network by becoming a sponsor. Any amount is appreciated!