Introduction
Through collaboration with EigenLayer, Cartesi has become the first and only blockchain technology stack capable of achieving on-chain fully trust-minimized large language model inference, with DeepSeek as a pioneering use case.
This marks a critical breakthrough in the convergence of cryptography and artificial intelligence, ushering in a new era where blockchains not only support AI, but can actively drive AI in a fully deterministic manner.
Background
Traditional AI relies on centralized servers and opaque "black box" systems, where users trust the outputs but cannot verify the underlying processes. This dependence on centralized infrastructure poses significant vulnerabilities: limited transparency, susceptibility to censorship or tampering, and an inevitable reliance on trust.
Many projects claim to offer "decentralized AI," but still rely on proprietary models, centralized infrastructure, and trust-based systems. As a result, most so-called decentralized AI practices remain constrained by traditional frameworks, merely recording AI outputs on the blockchain rather than using it as an active decentralized computing environment.
True decentralized AI must guarantee that the same inputs always produce the same outputs, and that anyone can verify this in a trustless environment. This paradigm requires provable determinism, where the AI computation itself runs on-chain, rather than just recording the results on the ledger.
Cartesi's Advantages in the AI Domain
Cartesi is the only project currently capable of achieving truly blockchain-native AI, with the unique ability to execute AI computations directly within smart contracts, ensuring complete transparency and verifiability.
Unlike traditional Trusted Execution Environments (TEEs), which enhance security but still have single points of failure, Cartesi provides a verifiable decentralized execution environment, enabling fully on-chain AI inference.
Cartesi's RISC-V virtual machine offers unique advantages, allowing developers to seamlessly run powerful AI libraries like TensorFlow, PyTorch, and Llama.cpp in a decentralized environment, breaking through the traditional limitations of blockchains in AI computation.
Cartesi's Collaboration with DeepSeek
During the recent Cartesi x Eigen experiment week, Cartesi's core engineering team successfully built a prototype system that enables smart contracts to execute verifiable large language model inference. This makes Cartesi the first project in the crypto space to achieve fully on-chain deployment of DeepSeek. The prototype system also supports the execution of other open-source LLMs, such as Qwen and Smol.
By executing inference within Cartesi's verifiable environment, the highest level of security is achieved by eliminating centralized intermediaries, ensuring the integrity of AI outputs while guaranteeing complete trustlessness. To date, Cartesi remains the only technology stack that supports fully trust-minimized on-chain LLM inference, with DeepSeek serving as a pioneering use case that validates its potential.
Future Outlook
The impact of decentralized AI goes far beyond the technical breakthrough itself. The emergence of truly deterministic AI will unlock unprecedented possibilities in domains such as autonomous treasury management, smart DeFi systems, and fully autonomous governance, reshaping the decentralized ecosystem.
About Cartesi
Cartesi is a powerful modular blockchain protocol that provides developers with a full Linux environment and high-performance Rollup technology, aiming to support the next generation of decentralized applications. By integrating Linux, the Cartesi Virtual Machine empowers developers to leverage the battle-tested programming languages, tools, and code libraries built over decades. Cartesi provides each dApp with a dedicated Rollup layer and specialized computing resources, significantly enhancing computational scalability while preserving decentralization, security, and censorship resistance.
