Lilypad 🍃

This cloud is just someone else's computer.

Read the docs!

This guide shows you (amongst other things) how to:

trigger jobs on other peoples' computers with just a metamask wallet and our CLI
trigger the same jobs from a smart contract deployed to our testnet
make crypto with your GPU by running Stable Diffusion fine-tuning & inference jobs for someone else.

Caveat: today it is just worthless test crypto, but one day soon it will be real crypto!

Get private key

Set up metamask with our Lalechuza testnet.

Click the down arrow on networks and click add network.

The variables you need are:

Network name: Lilypad Lalechuza testnet
New RPC URL: http://testnet.lilypadnetwork.org:8545
Chain ID: 1337
Currency symbol: ETH
Block explorer URL: (leave blank)

Click save and click "switch to Lilypad Lalechuza testnet".

Click the down arrow next to Account 1 on the top bar, then click the ... next to Account 1 in the list
Account details - then click Show private key

Show private key
Enter your metamask password
Copy private key for account - DO NOT COPY THE WALLET ADDRESS BY MISTAKE

Open a terminal on macOS, Linux or WSL2, and type:

export PRIVATE_KEY=<the private key you copied>

Go get some funds from the faucet: http://testnet.lilypadnetwork.org/

Now let's run a job!

Hello (cow) world example

Requires:

Docker

Works on Linux, macOS and WSL2 (x86_64 and arm64)

Install lilypad CLI:

curl -sSL -O https://raw.githubusercontent.com/bacalhau-project/lilypad-modicum/main/lilypad && sudo install lilypad /usr/local/bin/lilypad

run a job (x86_64 or arm64)

Run cowsay via the BLOCKCHAIN

export PRIVATE_KEY=<as above>
lilypad run --template cowsay:v0.0.1 --params "hello lilypad"

(ensure your user is in the docker group if necessary on your platform)

run a node (x86_64 only)

To contribute your resources to the network and get paid

export PRIVATE_KEY=<as above>
sudo -E lilypad serve

usage

stable diffusion (requires GPU)

lilypad run --template stable_diffusion:v0.0.1 --params "blue frog"

TODO:

fine tuning stable diffusion with LoRA
inference on a fine-tuned LoRA
t2i sketch, depth & pose
controlnet
support some subset of https://platform.stability.ai/docs/features

filecoin data prep

lilypad run --template filecoin_data_prep:v0.0.1 \
	--params '{"s3_bucket": "noaa-goes16", \
	           "s3_key": "ABI-L1b-RadC/2000/001/12/OR_ABI-L1b-RadC-M3C01*"}'

TODO: read results from http rather than ipfs for high performance

arbitrary wasm (run in a deterministic env)

TODO: the following seems to be a csv2parquet program that requires a CSV as input - need to also provide a CSV as input! (but it runs, giving the error message rn)

lilypad run --template deterministic_wasm:v0.0.1 \
	--params '{"wasm_cid": "Qmajb9T3jBdMSp7xh2JruNrqg3hniCnM6EUVsBocARPJRQ", \
	           "wasm_entrypoint": "_start"}'

eyeregens/lilypad0