{"id":"nesa","title":"Nesa","content":"**Nesa** is a Layer-1 blockchain network focused on decentralized AI execution, enabling AI models to run across distributed compute infrastructure with privacy-preserving and verifiable inference. The network provides tools for deploying decentralized AI applications (DAIs), managing AI models, and coordinating computation through cryptographic verification and blockchain-based mechanisms. [\\[1\\]](#cite-id-wRg24h1WuoKuZMI7) \n\n## Overview\n\nNesa is a Layer-1 blockchain designed to support decentralized AI execution with a focus on privacy, verification, and distributed computation. Rather than relying on centralized AI providers, the network enables AI inference across a distributed network of compute nodes while keeping both the underlying models and user data confidential. The protocol uses cryptographic techniques, including Equivariant Encryption, Homomorphic Secret Sharing over Encrypted Embeddings (HSS-EE), and the MetaInf inference scheduler, to verify off-chain AI computation on-chain and optimize execution across different hardware environments. It also includes research into scalable model sharding for distributed AI processing.\n\nThe network supports decentralized AI (DAI), enabling developers to build AI-powered applications that run natively on Nesa without managing their own infrastructure. AI requests are encrypted, distributed across multiple nodes for execution, and returned with cryptographic proofs verifying the correctness of the results. Nesa also provides a decentralized marketplace for AI models, supporting both open-source and proprietary models across multiple AI modalities, as well as a model upload pipeline and execution environment. Its architecture is designed to allow a broader range of nodes to participate through model sharding and distributed coordination, reducing reliance on specialized hardware while supporting governance, incentives, and standardized deployment within the protocol. [\\[2\\]](#cite-id-6HRVJZmlOOUN1KjL) \n\n![](https://ipfs.everipedia.org/ipfs/QmZjexXk33SXUnv8cqaQw3uaWi3XPe7TwMi3LuJBaUkw94)\n\n## Technology\n\n### Decentralized AI\n\nNesa's decentralized AI execution framework enables AI inference requests to be processed across a distributed network of nodes rather than by a single centralized provider. The system uses a two-phase commit-reveal process to discourage dishonest behavior and free-riding, while smart contracts verify and aggregate inference results on-chain. This architecture is designed to provide transparent, verifiable AI computation while distributing workloads across network participants.\n\nThe platform also supports Decentralized AI Applications (DAIs), which are AI products built on Nesa's infrastructure. Developers can deploy AI applications without building their own blockchain integration, decentralized inference, or incentive systems, as Nesa provides the underlying execution, coordination, and management framework. DAIs can access features such as subsidized inference during early deployment, platform-based discovery, token issuance on Nesa's EVM testnet, and staking mechanisms that allow users to signal confidence in an application's long-term development. These components are intended to support the deployment, operation, and governance of AI applications within Nesa's decentralized ecosystem. [\\[4\\]](#cite-id-YPsjRyi0dqQ2Z8Nn) \n\n#### Model Playground\n\nNesa provides a Model Playground that serves as a catalog for exploring, evaluating, and deploying AI models on its decentralized network. The catalog includes models for text, image, video, audio, multimodal, and domain-specific tasks, with metadata such as supported workloads, inference latency, usage metrics, and deployment status. These models run on Nesa's decentralized inference infrastructure, allowing developers to compare and integrate them into decentralized AI applications (DAIs) or other software without managing hosting or execution infrastructure.\n\nIn addition to the public model catalog, Nesa supports deploying custom AI models via a model upload pipeline available to Pro users. Developers can register models using Docker images or GitHub repositories, define model metadata, task types, and access settings, and deploy them as standardized network assets. Uploaded models are compatible with Nesa's decentralized execution framework and can be integrated with DAIs, staking, and protocol incentive mechanisms while maintaining consistent execution and discovery across the network. [\\[5\\]](#cite-id-nTTS1nYvYGcqM6D6) \n\n### Equivariant Encryption (EE)\n\nEquivariant Encryption (EE) is Nesa's encryption framework for performing AI inference on encrypted data while maintaining performance comparable to standard AI execution. Instead of relying on trusted hardware or computationally intensive cryptographic methods such as fully homomorphic encryption, EE transforms AI models to operate directly on encrypted inputs. The framework is designed so that encrypted inference produces the same results as plaintext inference after decryption, while keeping user inputs, intermediate computations, and outputs encrypted throughout execution.\n\nEE is intended to support modern AI architectures, including large language models, computer vision models, retrieval-augmented generation (RAG) systems, and multimodal models, with minimal changes to existing model implementations. According to Nesa, benchmark testing showed less than 9% inference latency overhead and output fidelity exceeding 99.99% compared with standard inference. The framework is designed to operate on conventional GPU hardware without trusted execution environments, and can be extended through HSS-EE, which combines Equivariant Encryption with Homomorphic Secret Sharing to distribute computation across multiple non-colluding servers while preserving encrypted inference and reducing trust in any single execution node. [\\[3\\]](#cite-id-EPAue8g6NNPVYWjL) \n\n## NES\n\nThe $NES token is the native utility token of the Nesa network and is used to pay transaction fees, participate in governance, support network security, and coordinate decentralized AI inference. It has a total supply of 1B tokens. Developers submit **PayForQuery** transactions when requesting AI inference or publishing query data to the network, with transaction fees determined by a combination of a fixed base fee and a variable fee proportional to the request size. Validators prioritize transactions through a gas-price-based fee market, while transaction data is processed, verified, and recorded on-chain using erasure coding, Merkle trees, and Byzantine Fault Tolerant (BFT) validation mechanisms. Nesa's **queryStream** system can also relay verified inference data to external settlement layers, including Ethereum.\n\nThe token also plays a role in governance and network incentives. NES holders can propose and vote on changes to selected protocol parameters, while a portion of block rewards is allocated to a community pool that funds ecosystem initiatives through governance decisions. Network participants responsible for AI query processing are required to stake NES, creating economic incentives for honest behavior and enabling slashing penalties for malicious activity. According to Nesa's design, the amount of NES committed can influence the number of participating validators or processing nodes that verify AI inference, allowing users to balance transaction costs with the desired level of verification and security. [\\[6\\]](#cite-id-lLVQJjeSyBE3Zg3X) \n\n![](https://ipfs.everipedia.org/ipfs/QmegT9q6BKSgK3fBLiPzR1mifwyXtaZNCxEp1vjPA5k1gP)\n\n## Partnerships\n\n* SecondLive\n* Flock\n* Story\n* Holoworld\n* Privasea\n* XION\n* Manta Network\n* BNB Chain\n* Sei","summary":"Nesa is a privacy-preserving Layer-1 blockchain enabling decentralized AI execution with tools for deploying and managing decentralized AI applications.","images":[{"id":"QmWMLXEKagnG8yWaMXBXtZxTevAm5FhCMuN6PBLqmUZG4g","type":"image/jpeg, image/png"}],"categories":[{"id":"projects-and-protocols","title":"projects-and-protocols"}],"tags":[{"id":"Layer1"},{"id":"AIPlatform"},{"id":"AIInfrastructure"},{"id":"Ethereum"}],"media":[{"id":"QmWMLXEKagnG8yWaMXBXtZxTevAm5FhCMuN6PBLqmUZG4g","name":"nesalogo.png","caption":"","thumbnail":"QmWMLXEKagnG8yWaMXBXtZxTevAm5FhCMuN6PBLqmUZG4g","source":"IPFS_IMG"},{"id":"QmZjexXk33SXUnv8cqaQw3uaWi3XPe7TwMi3LuJBaUkw94","name":"Nesa4.png","caption":"","thumbnail":"QmZjexXk33SXUnv8cqaQw3uaWi3XPe7TwMi3LuJBaUkw94","source":"IPFS_IMG"},{"id":"QmVrv8deBnWuR9aFtChNRNPF25aiB9SnTC7mQPo7X9BGJH","name":"Nesa3.png","caption":"","thumbnail":"QmVrv8deBnWuR9aFtChNRNPF25aiB9SnTC7mQPo7X9BGJH","source":"IPFS_IMG"},{"id":"QmegT9q6BKSgK3fBLiPzR1mifwyXtaZNCxEp1vjPA5k1gP","name":"Nesa2.png","caption":"","thumbnail":"QmegT9q6BKSgK3fBLiPzR1mifwyXtaZNCxEp1vjPA5k1gP","source":"IPFS_IMG"},{"id":"Qmdc5G9bMQfFZsKMYmdDogMvzqNCmdmAvmb1k33NZoVSVW","name":"Nesa1.png","caption":"","thumbnail":"Qmdc5G9bMQfFZsKMYmdDogMvzqNCmdmAvmb1k33NZoVSVW","source":"IPFS_IMG"}],"metadata":[{"id":"references","value":"[\n {\n \"id\": \"wRg24h1WuoKuZMI7\",\n \"url\": \"https://nesa.ai\",\n \"description\": \"Nesa official site\",\n \"timestamp\": 1783688163611\n },\n {\n \"id\": \"6HRVJZmlOOUN1KjL\",\n \"url\": \"https://docs.nesa.ai/nesa\",\n \"description\": \"Nesa Docs\",\n \"timestamp\": 1783688163611\n },\n {\n \"id\": \"EPAue8g6NNPVYWjL\",\n \"url\": \"https://docs.nesa.ai/nesa/major-innovations/private-inference-for-ai/practical-approach-equivariant-encryption-ee\",\n \"description\": \"Equivariant Encryption Details\",\n \"timestamp\": 1783688163611\n },\n {\n \"id\": \"YPsjRyi0dqQ2Z8Nn\",\n \"url\": \"https://docs.nesa.ai/nesa/major-innovations/decentralized-execution-for-ai/dais-building-decentralized-ai-applications-on-nesa\",\n \"description\": \"Building Decentralized AI Applications on Nesa\",\n \"timestamp\": 1783688163611\n },\n {\n \"id\": \"nTTS1nYvYGcqM6D6\",\n \"url\": \"https://docs.nesa.ai/nesa/major-innovations/decentralized-execution-for-ai/models-on-nesa-playground-and-custom-uploads\",\n \"description\": \"Models on Nesa: Playground and Custom Uploads\",\n \"timestamp\": 1783688163611\n },\n {\n \"id\": \"lLVQJjeSyBE3Zg3X\",\n \"url\": \"https://docs.nesa.ai/nesa/major-innovations/usdnes-utility\",\n \"description\": \"Utility of $NES\",\n \"timestamp\": 1783688163611\n },\n {\n \"id\": \"mvcaKhYKj4B6yoDo\",\n \"url\": \"https://www.rootdata.com/projects/detail/Nesa?k%3DMTE2MDA%3D\",\n \"description\": \"Project Partners\",\n \"timestamp\": 1783688163611\n },\n {\n \"id\": \"UMcJ1qAIlCq2FOi1\",\n \"url\": \"https://x.com/nesaorg\",\n \"description\": \"Nesa on X\",\n \"timestamp\": 1783688163611\n }\n]"},{"id":"website","value":"https://nesa.ai"},{"id":"twitter_profile","value":"https://x.com/nesaorg"},{"id":"discord_profile","value":"https://discord.com/invite/Z4jZPTdfcY"},{"id":"explorer_injective_profile","value":"https://explorer.nesa.ai/"},{"id":"rootdata_profile","value":"https://www.rootdata.com/projects/detail/Nesa?k=MTE2MDA%3D"},{"id":"github_profile","value":"https://github.com/nesaorg/nesa"},{"id":"telegram_profile","value":"https://discord.gg/nesa"},{"id":"medium_profile","value":"https://nesa.ai/research"},{"id":"bscscan","value":"https://bscscan.com/address/0x3131f6B80C26936aB03F7d9D29Eb4Ddf36AC3FB5"},{"id":"nansen","value":"https://app.nansen.ai/token-god-mode?chain=ethereum&tab=transactions&tokenAddress=0x230f1E241C621d5af670Dad83ebCdd18971E2995"},{"id":"etherscan","value":"https://etherscan.io/token/0x230f1E241C621d5af670Dad83ebCdd18971E2995"},{"id":"coinmarketcap_url","value":"https://coinmarketcap.com/currencies/30140/"},{"id":"commit-message","value":"\"Added new wiki Nesa\""}],"events":[{"id":"0bb64a8c-cdbd-4455-b2d0-2b85c281e086","date":"2023-06-01","title":"Nesa Launched","type":"CREATED","description":"Launch of Nesa blockchain focusing on decentralized AI execution.","link":null,"continent":null,"country":null}],"user":{"id":"0x8af7a19a26d8fbc48defb35aefb15ec8c407f889"},"author":{"id":"0x8af7a19a26d8fbc48defb35aefb15ec8c407f889"},"operator":{"id":"0xacb6c5AD52b8f605299B0d774CE97F26e3DB80c2"},"language":"en","version":1,"linkedWikis":{"blockchains":["ethereum"],"founders":["anon"],"speakers":[]}}