Gensyn AI

The Gensyn Protocol is a layer-1 trustless protocol for deep learning computation that directly and immediately rewards supply-side participants for pledging their compute time to the network and performing ML tasks. The protocol requires no administrative overseer or legal enforcement, rather facilitating task distribution and payments programmatically through smart contracts.

Total Raised $50.60M from a16z, Eden Block, CoinFund, Maven 11 Capital, Hypersphere Ventures, 7percent Ventures and other funds.

Statement from the official litepaper: 

1. Purchase the node and wait until it is in 'deployment' state
2. Open your node page and the follow the link: 'Check your node'
3. Log in to the Gensyn node dashboard using email (google auth doesn't work now). Use unique email for each node or you will be logged out everywhere! Once you see "YOU ARE SUCCESSFULLY LOGGED IN TO THE GENSYN TESTNET" this means the setup is COMPLETED and node is working, no need to do anything else or contact us. Please don't close the page immediately and wait for at least 10-15 minutes with the page opened so that the node will initialize correctly!
IMPORTANT: If you see the "Your connection is not private" warning in your Google Chrome, just ignore it. Click Advanced -> Proceed and the page will open.
4. Use node name to check the Gensyn dashboard to track your node: https://dashboard.gensyn.ai/
Please note that right now Gensyn Dashboard only shows top 100 nodes, so if even your node is online and healthy, you might not see it in the list.
6. Use this website to track your wins: https://gensyn-node.vercel.app/
7. Copy your "Connected EOA address" and past it in the explorer: https://gensyn-testnet.explorer.alchemy.com/
If you see internal transactions coming daily then everything is good.
8. You can also use unofficial TG bot: @gensyntrackbot to track your node. Please wait at least 2-3 days before your points are displayed as the network is delayed.

Q:  Why is my EVM address in nodes.garden is different from the EVM address in @gensyntrackbot?
A: If addresses are different - it's fine and everything works as expected. Reponse from the Gensyn team:
"we use alchemy smart accounts for account abstraction, so the correct accounts are the ones you can see onchain - not the addresses in the local JSON files (those are the smart accounts)."
But, if you see 0x0000000000000000000000000000000000000000 in the bot address, that means your node didn't initialize properly and contribution is not being recorded. This is an issue on Gensyn side and we cannot fix it. But we can make a fresh setup for you and hope it initializes correctly. Please contact us and provide Node ID from nodes.garden so we can do it quick.

Q: Why am I waiting for a few days and have 0 wins in the tracking bot?
A: The CPU nodes are less performant than GPUs, so amount of wins will be significantly less than GPU nodes. You might not get any wins for days on CPU. YOU ARE NOT GUARANTEED TO GET WINS OR TRANSACTIONS EVERY DAY IF YOU USE CPU NODE. If you want to farm a lot of wins please don't choose CPU node. If your CPU node doesn't get any wins/transactions in 2-3-4-5 days this is absolutely normal, there is nothing to do or to fix. If your CPU node has 0 wins after one week then please contact us so we can take a look.

Additional comment from the Gensyn team:
"If you address is doing transactions on chain and you can see that in the explorer then everything is fine. Amount of wins is less important."

Q: I see BlockingIOError: [Errno 11] Resource temporarily unavailable — is my node broken?
A: No need to panic — this is expected behavior. Gensyn nodes are currently critically unstable, especially on CPU configurations. These errors are part of the normal operation and cannot be fixed or prevented. The node will automatically refresh and recover during the next round (usually within a few hours).
⚠️ Please do not contact support about this issue, as there is nothing we can do to resolve it. Thank you for your understanding and patience.

Q:  Which model do you use? Can you run 7B hard math model on GPU?
A: We use 0.5b math model for both CPU/GPU. 1.5B model is extremely unstable and 7B model requires at least 80GB vRAM making it impossible to run on RTX3090/4090. So no, we can not run 1.5B or 7B models.

English


Hindi


Hindi