We are thrilled to partner with Marlin, the leading verifiable computation layer. Through this collaboration, smart agents in Talus will have access to secure compute verified by onchain proofs and attestations.

Verifiable computation entails data processing with verifiable proofs. In AI, this creates the possibility of training models and deploying smart agents without revealing the inputs themselves and creating verifiable outputs. This enables decentralized and censorship-resistant AI applications that were previously impossible onchain.

Talus is fully committed to this vision. Through its AI Smart Agent Hub, developers can access resources and customize levels of privacy and verifiability as needed for each specific scenario. Users will benefit from cryptographic assurances that their data is kept safe and not abused by third parties.

This is where Marlin comes in. Marlin’s co-processor, combined with a vast array of GPU/CPU clusters and privacy-preserving mechanisms, enables cheap, scalable compute on private data. With a Trusted Execution Environment (TEE), developers can access isolated and encrypted environments for the most sensitive workloads. This enables smart agents to inherit these properties to perform tasks without revealing sensitive information.

Here’s how it works:

1. Developer accesses Marlin compute infrastructure via the Talus Smart Agent Hub.

2. Developer chooses his GPU/CPU preference from Marlin.

3. Developer opts for TEE.

4. Developer deploys the Smart Agent on Talus.

5. Smart agent does an AI inference within a TEE

6. Smart agent uses verifiable inference to perform an action.

Looking Ahead

Together, Talus and Marlin are pioneering a new wave of verifiable compute for smart agents. Developers will utilize privacy-preserving mechanisms to perform computations on sensitive data, enhancing user trust and powering more complex, advanced agentic activity across the crypto economy.

Talus is building the universal deployment engine for AI Smart Agents.

We are creating a decentralized protocol where individuals can freely exchange data, models, and computational power to make AI transparent and verifiable. By leveraging the security, performance, and developer experience of Move, Talus is the foundation of the Smart Agent Revolution and a new era of consumer AI.

To learn more about Talus, follow us on Twitter, check out our GitHub, visit our Discord, and read our Mirror.

Marlin is a verifiable computing protocol featuring TEE and ZK-based coprocessors to delegate complex workloads over a decentralized cloud. Servers provisioned using smart contract calls host ML models, gateways, frontends, MEV or automation bots, or backends for arbitrary computations using external APIs with baked-in auto-scaling and fault tolerance. Marlin is backed by Binance Labs and Electric Capital.

To learn more about Marlin, follow them on Twitter, check out their GitHub, visit their Discord and read their Blog.

We are thrilled to partner with Marlin, the leading verifiable computation layer. Through this collaboration, smart agents in Talus will have access to secure compute verified by onchain proofs and attestations.

Verifiable computation entails data processing with verifiable proofs. In AI, this creates the possibility of training models and deploying smart agents without revealing the inputs themselves and creating verifiable outputs. This enables decentralized and censorship-resistant AI applications that were previously impossible onchain.

Talus is fully committed to this vision. Through its AI Smart Agent Hub, developers can access resources and customize levels of privacy and verifiability as needed for each specific scenario. Users will benefit from cryptographic assurances that their data is kept safe and not abused by third parties.

This is where Marlin comes in. Marlin’s co-processor, combined with a vast array of GPU/CPU clusters and privacy-preserving mechanisms, enables cheap, scalable compute on private data. With a Trusted Execution Environment (TEE), developers can access isolated and encrypted environments for the most sensitive workloads. This enables smart agents to inherit these properties to perform tasks without revealing sensitive information.

Here’s how it works:

1. Developer accesses Marlin compute infrastructure via the Talus Smart Agent Hub.

2. Developer chooses his GPU/CPU preference from Marlin.

3. Developer opts for TEE.

4. Developer deploys the Smart Agent on Talus.

5. Smart agent does an AI inference within a TEE

6. Smart agent uses verifiable inference to perform an action.

Looking Ahead

Together, Talus and Marlin are pioneering a new wave of verifiable compute for smart agents. Developers will utilize privacy-preserving mechanisms to perform computations on sensitive data, enhancing user trust and powering more complex, advanced agentic activity across the crypto economy.

Talus is building the universal deployment engine for AI Smart Agents.

We are creating a decentralized protocol where individuals can freely exchange data, models, and computational power to make AI transparent and verifiable. By leveraging the security, performance, and developer experience of Move, Talus is the foundation of the Smart Agent Revolution and a new era of consumer AI.

To learn more about Talus, follow us on Twitter, check out our GitHub, visit our Discord, and read our Mirror.

Marlin is a verifiable computing protocol featuring TEE and ZK-based coprocessors to delegate complex workloads over a decentralized cloud. Servers provisioned using smart contract calls host ML models, gateways, frontends, MEV or automation bots, or backends for arbitrary computations using external APIs with baked-in auto-scaling and fault tolerance. Marlin is backed by Binance Labs and Electric Capital.

To learn more about Marlin, follow them on Twitter, check out their GitHub, visit their Discord and read their Blog.