Run a Lilypad job from a FVM smart contract. The Lalechuza testnet is built on geth & allows arbitrary untrusted nodes to join the network. It works with “modules” which are Lilypad jobs (currently deterministic-only).

Modules include Filecoin data prep, Stable Diffusion (SDXL0.9), LoRA fine tuning module, LLM Inference, DuckDB, and more! Improved developer experience with detailed documentation and examples.

Anyone can contribute compute to the network. The testnet has a base currency of ETH & a utility token called LP. Receive LP to pay for jobs (and nodes to stake) from the Lilypad faucet .v2 is an IPC compatible network written in Go.

Platform for users to freely access leading generative AI models such as Stable Diffusion XL (SDXL) and Mistral-7B-Instruct through an intuitive web interface.

Devs and infra providers contribute their machines and experiment with running workloads directly over decentralized models and datasets. The goal is to stress test core network functions like job scheduling, compute resource allocation and data routing.

Gather and implement feedback from intital Calibration Net Resource Providers ensuring a smooth transition to Incentivized Testnet. Add a variety of GPUs to run many different sized workloads.

The Incentivized testnet will run in phases testing out and rewarding different aspects of the market including compute providers, module providers, security bounties, game theory aspects, and application developers.

Continual refinement and enhancing the testnet with integration componentry such as SDKs and APIs to ensure a network that is thriving and highly extensible. Alongside this, ensuring the right partnership alignments to bring cross collaboration and enhanced job workload capabilities to the network.

Mainnet launch with Liliypad token (LP) paid to compute providers.