Fuel up
Pull completed candles from Binance, Coinbase or Hyperliquid and turn them into signals and yes/no answers.
GL1FCryptopowered by GenesisL1Early alpha
No-code crypto AIRuns in your browser · GenesisL1 EVM · Chain 29
Launch your own Crypto AI model.
Pick a coin. Ask a yes-or-no question about the next few hours or days. Your Crypto AI model studies the candles for an answer, or tells you honestly there isn’t one. Backtest it with fees, mint it on GenesisL1 as an on-chain Crypto AI model, share it. No code, no server, no sign-up.
Educational, experimental software. Not investment advice. Trading can lose you everything. No Lambo guaranteed. Read the disclaimer.
01/ Questions your Crypto AI model can answer
Ask your Crypto AI model, from majors to the trenches.
Every card opens the studio with the settings filled in. The answer can be “no edge here”, and that is worth knowing before you risk anything.
ETHBinance · 15m
Will ETH rise 1% before it drops 0.5% in the next 5 hours?
Up 1%Stop 0.5%5h
Build this model →BTCBinance · 1h
Is BTC about to push 1.5% higher before a 1% pullback, within half a day?
Up 1.5%Stop 1%12h
Build this model →SOLBinance · 1h
When funding runs hot, does SOL cool off 2% before squeezing 1% higher?
Down 2%Stop 1%8h
Build this model →ETH-USDCoinbase · 5m
On Coinbase spot, does ETH-USD gain 0.8% before losing 0.4% within 2 hours?
Up 0.8%Stop 0.4%2h
Build this model →XRPBinance · 4h
Does XRP slide 3% before it bounces 1.5% over the next day?
Down 3%Stop 1.5%24h
Build this model →LINKBinance · 1h
Is LINK trending cleanly enough to add 2% before giving back 1%?
Up 2%Stop 1%12h
Build this model →DOGEBinance · 15m
Will DOGE send 5% before it dumps 2% in the next 6 hours?
Up 5%Stop 2%6h
Build this model →1000PEPEBinance · 5m
Is this PEPE volume spike the start of a run, or the top?
Up 4%Stop 2%2h
Build this model →WIFBinance · 15m
Does WIF hold a pump into the New York session?
Up 3%Stop 1.5%4h
Build this model →1000BONKBinance · 15m
After a vertical candle, does BONK retrace 6% before squeezing 3% higher?
Down 6%Stop 3%8h
Build this model →DOGEBinance · 15m
When BTC dumps, do memes dump harder? DOGE −3% before +1.5%, within 4 hours.
Down 3%Stop 1.5%4h
Build this model →1000SHIBBinance · 1h
Can SHIB do +3% before −1.5% by tomorrow, or is it just vibes?
Up 3%Stop 1.5%24h
Build this model →HYPEHyperliquid · 15m
Does HYPE rip 3% before it dips 1.5% within 6 hours?
Up 3%Stop 1.5%6h
Build this model →kPEPEHyperliquid · 15m
Can kPEPE send 5% before a 2.5% flush in the next 8 hours?
Up 5%Stop 2.5%8h
Build this model →ETHHyperliquid · 1h
With hourly funding in the mix, does ETH add 1.5% before losing 1%?
Up 1.5%Stop 1%12h
Build this model →BTCHyperliquid · 1h
Does BTC drop 1% before it pops 0.5% within 6 hours?
Down 1%Stop 0.5%6h
Build this model →SOLHyperliquid · 15m
Is SOL’s momentum good for +2% before −1% within 4 hours?
Up 2%Stop 1%4h
Build this model →DOGEHyperliquid · 15m
Will DOGE pump 4% before it dumps 2% in 6 hours on Hyperliquid?
Up 4%Stop 2%6h
Build this model →02/ Explore
Crypto AI models already on-chain.
Published GL1F Crypto AI models, read live from GenesisL1. Open one in the studio to run it on the latest candle and verify the score on-chain.
03/ How it works
Five laps to your own Crypto AI model.
Everything runs in your browser tab. Candles come straight from the exchange’s public API; only deploying sends a transaction.
Tune the engine
Gradient-boosted trees learn from the past and are tested walk-forward on the future.
Test drive
Replay history with fees, slippage and stops: equity curve, drawdown, win rate.
Get the keys
Mint the Crypto AI model as a Crypto AI Model NFT on GenesisL1. Protocol fees are burned by anyone.
Hit the road
Run it on the latest candle or any moment in history, and verify the score on-chain.
04/ Model monetization
Your Crypto AI model, your business model.
The owner of a Crypto AI Model NFT is its model admin and decides how it earns. The admin can change it any time, or sell the model on the marketplace; the buyer becomes the new admin.
Open to everyone
Anyone can run the model and verify its scores on-chain. Good for reputation, research and teaching.
Pay what you like
Inference stays open, and users can tip the fee recipient when a model taught them something.
A fee per run
Every on-chain inference pays the fee the admin sets, straight to the admin's fee recipient.
Time-based access
Sell plans such as 30 days for 1 L1. A subscriber pays once, and their access key runs the model until the plan ends.
Crypto AI model marketplace
List your Crypto AI Model NFT at your price, buy someone else's, or cancel a listing. A sale moves the NFT and, with it, the admin rights: access, fees and subscription plans.
Who gets paid
Inference fees, tips, subscriptions and sale prices go to the model admin or seller. Protocol fees (creation, per-byte storage, listing) are burned by anyone and can never be withdrawn.
05/ The machine
GenesisL1 EVM is your quant.
Your Crypto AI model is a GL1F gradient-boosted tree ensemble whose integer inference returns exactly the same score in your browser and inside the EVM. GenesisL1, an EVM Layer 1 live since 2021, stores it, prices it and runs it, so anyone can recompute a prediction on-chain. JavaScript, Python and C++ trainers emit identical Crypto AI model cores on the published test matrix. The chain does the quant work; you ask the questions.
See the contracts- Engine
- GL1F integer trees
- Same integer score in your browser and the EVM.
- Chassis
- GenesisL1 · chain 29
- EVM Layer 1 since 2021.
- Fuel
- Completed candles
- Binance and Hyperliquid perpetuals, Coinbase spot. Never the candle still forming.
- Transmission
- Walk-forward validation
- Train on the past, test on the next slice of time. 2 to 20 folds, your call.
- Brakes
- Backtest with costs
- Fees, slippage, stops. Most “amazing” Crypto AI models stop here. That’s the point.
- Keys
- Crypto AI Model NFT
- Your Crypto AI model, your token: free, tips, pay per inference or subscription plans. The owner is the model admin.
- Exhaust
- Burned fees
- Creation, per-byte and listing fees go to 0x…dEaD. Anyone can press the button.
29Chain id
212Causal signals
3Exchanges
2–20Walk-forward folds
0Withdrawable fees
06/ Learn
AI for crypto traders, in five minutes.
Six ideas that separate a useful Crypto AI model from a pretty chart. The studio explains each one again, right where you need it.
Features are the eyes
Signals the Crypto AI model can see at each candle: recent returns, volatility, volume surges, funding, how BTC is moving. Only past data, never the future.
The label is the question
For every candle: did price hit your target before your stop, within your window? 1 or 0. The Crypto AI model learns to estimate that probability.
Trees vote together
Hundreds of small decision trees each fix the mistakes of the ones before. Together they form a probability, not a promise.
Overfitting is the enemy
A Crypto AI model that memorised last month looks brilliant on last month and useless tomorrow. Early stopping and simple settings fight it.
Test on the future
Walk-forward validation trains on the past and tests on what came next. Shuffling time together flatters the Crypto AI model and fools you.
70% means wrong 3 in 10
Probabilities are frequencies, not certainties. After fees, a small edge can vanish entirely. The backtest tells you whether it did.
07/ FAQ
Questions about crypto AI.
What is a Crypto AI model?
A machine-learning model that reads market signals such as recent returns, volatility, volume, funding and how Bitcoin is moving, and outputs the probability that your yes-or-no question comes true, for example: will ETH rise 1% before it drops 0.5% in the next 5 hours? GL1F Crypto builds it from exchange candles with gradient-boosted decision trees.
Can AI predict crypto prices?
Sometimes a little, often not at all. Markets are noisy, and most patterns are weak or vanish after fees. That is why every Crypto AI model here is tested walk-forward on data it never saw and backtested with costs, and why the honest answer can be: no edge.
Is this an AI trading bot?
No. GL1F Crypto is an educational Crypto AI model builder. It never places trades, never connects to exchange accounts or funds, and its Crypto AI models must not be used for live or automated trading.
What makes it on-chain AI?
The Crypto AI model is stored on GenesisL1 as an Crypto AI Model NFT, and its inference runs as integer math inside the EVM, so anyone can recompute a prediction on-chain and get exactly the score your browser shows. GenesisL1 EVM is your quant.
What is GL1F Crypto?
A free, open-source crypto AI studio that turns a market question into a machine-learning Crypto AI model. It downloads completed candles from Binance, Coinbase or Hyperliquid, trains gradient-boosted trees in your browser, backtests them with fees, and can mint the Crypto AI model as an Crypto AI Model NFT on GenesisL1 so anyone can verify its predictions.
Do I need to code or know machine learning?
No. Every field has a question mark with a plain-language explanation, and the defaults are sensible. You pick a coin, a target move, a stop and a time window; the studio does the rest.
Will it make me money?
Nobody can promise that, and we don't. GL1F Crypto is educational, experimental software. Most market patterns are weak and many disappear after fees. Crypto AI models must not be used for live trading. Crypto trading can lose you all of your money.
Where does my data go?
Nowhere. Candles are fetched straight from the exchange's public API into your browser, and training runs locally. Only deploying a model sends a transaction to GenesisL1.
What happens to deployment fees?
The model creation fee and the per-byte storage fee stay in the registry contract, and anyone can burn them to an unspendable address with one public call. No one, including the deployer, can withdraw them.
Why does my Crypto AI model look great in training and bad in the backtest?
That is usually overfitting, fees, or both, and it's exactly what the backtest is for. Try a simpler Crypto AI model (fewer trees, lower depth), turn on heuristic search, pick a bigger target or a longer candle, or accept that this question has no edge.
Why does Hyperliquid start at 15-minute candles?
Its public API keeps only the latest 5,000 candles per size. The signals need 22 days of warm-up history, which smaller candles can't cover, so the studio starts at 15m and sets the earliest possible start date for you.
Can I use someone else's Crypto AI model?
Yes. Published Crypto AI models can be run on the latest candle or any past moment and their scores verified on-chain; paid ones offer a fee per run or a subscription. Each shows the license it was published under.
Which license does a Crypto AI model NFT have?
The one its creator picked when minting it, from a catalog of standard licenses stored on GenesisL1: public domain (such as CC0), permissive (such as MIT, Apache 2.0, CC BY 4.0 or OpenMDW, made for machine-learning models), share-alike (such as CC BY-SA 4.0, GPL or ODbL) or restricted (non-commercial or no changes). The default is CC BY-SA 4.0. The license is written on-chain at mint; later its admin can only open it up or move it to a later version, even after a sale.
Can someone copy a model from the chain?
Technically yes: every model's bytes are public on GenesisL1. Legally it depends on the model's license. Open licenses allow copying; the GL1F On-Chain Use License, preselected for paid models, reserves all rights and allows use only through GenesisL1 on the admin's terms. The Terms of Service bind every user of the app and the contracts to respect each model's license.
Read this before you touch anything.
- Open, experimental data-science software for education and research only. It is not a trading product, a signal service, a broker or an adviser.
- No investment advice. Nothing here is financial, investment, trading, legal or tax advice, or a recommendation to buy, sell or hold any asset.
- Trading crypto carries serious risk, including the loss of everything you put in. Leverage multiplies losses. Memecoins can go to zero in minutes.
- Crypto AI models and backtests can be wrong, overfit, stale or broken. Past performance does not predict future results. A backtest is a simulation, not a record of real trades.
- Do not use these Crypto AI models in live trading, bots, copy-trading or any automated setup.
- No warranty of any kind. The software, website, smart contracts, Crypto AI models and data are provided “as is” and “as available”, without warranty of any kind, express or implied, including merchantability, fitness for a particular purpose, accuracy, availability and non-infringement.
- Bugs, errors and imperfections. This software may contain bugs, errors, defects, inaccuracies and other imperfections, and may not work as described. The smart contracts are not audited. Market data comes from third parties and may be wrong, delayed or incomplete.
- No liability. To the fullest extent permitted by law, the authors, contributors and affiliated parties are not liable for any loss or damage of any kind, including lost funds or profits, arising from the use of, or inability to use, the software or its results.
- On-chain actions are permanent. Deploying a Crypto AI model is a public transaction; fees are burned and never refunded.
You are solely responsible for your decisions and for complying with the laws of your jurisdiction. Exchange names identify public data sources only; GL1F Crypto is not affiliated with Binance, Coinbase or Hyperliquid.