The Real DeAI Anyone can Join and Use Right Away!
"The key to the real DeAI is a robust consensus protocol that enables the permissionless joining and using of the decentralized network by millions. The ability to identify and penalize all the malicious behaviors ensures the ecosystem's sustainability and facilitates healthy growth in a long term"
Based on the consensus protocol, the latest release, Helium Network, allows everyone to join, providing GPU to execute both the Stable Diffusion and GPT tasks for other applications.
Read More...
Application Developers
Use the network as an inference API service on the cloud to execute the Stable Diffusion and GPT tasks.
Get Started >>>
Image Generator
A demo of an application using the Helium Network. Generate images in the browser with no GPU.
Try it Now >>>
Start a Node
Connect your device to the network in exchange for tokens at idle time. NVIDIA GPU and Mac are both supported. The node could be started in seconds.
Get Started >>>
Use case for applications, AI model shop. Models, such as the anime diffusion model and legal document writer model, could be listed for trading. Models could be used as the API service and downloaded for local deployment.
For Applications
  • There is always a model fits your need in the marketplace.
  • Bring AI power to your applications simply using API.
  • Easy model customization using local data.
Marketplace (opening soon)  →
Use case for developers. Program the training task the usual way and start training on the data and computing power provided by the network. Online IDE and debugging tools.
For Developers
  • Utilize the dataset and computing resources provided by the network.
  • Write traditional PyTorch and Pandas tasks. Transparently run it in the decentralized network.
  • Developer friendly framework, SDK and debugging tools.
Developer Console  →
Use case for nodes. Provide private data to the network as datasets at no leakage risk. Decentralized network to join and quit at any time.
For Nodes
  • Earn from your private data and spare computing resources.
  • Private data is protected from public access with Federated Learning and MPC.
  • Transactions are verified using Zero Knowledge Proofs to prevent cheating.
Node Hosting Guide  →
Tech Blog View More
A Practical Verification Mechanism for Decentralized GPU Sharing
July 21, 2023
Stable Diffusion Workflow Reimagined: Cheap Flexible and more Powerful
June 18, 2023
LLM Fine-tuning using the Distributed Private Data
June 7, 2023
On-chain Verification of the LR Model Training using ZKP
Oct 26, 2022
It's a Fight for the Open and Democratic AI.
Blog Documentation GitHub
Contact us Twitter Discord
© 2023 Crynux. All Rights Reserved.