We are excited to announce our new collaboration with the Allora Network, working together to build the decentralized ecosystem for self-improving AI agents. This partnership will speed the development of AI agents for a massive range of onchain and offchain use cases.

Onchain AI smart agents will shape the future of app development and how we use blockchains. Allora's self-improving intelligence helps our network enable that intersection.

The partnership between Talus and Allora is built on a strong foundation of complementary technologies and shared vision. Allora provides the inferences and self-improvement mechanisms, while Talus provides the models and architecture for developers.

Allora brings machine intelligence to Talus through a pluggable version of their models. Developers can now leverage the network's machine intelligence through Move modules. While the models are stored offchain, their ownership and monetization are managed onchain on Talus. This unique approach allows developers to create objects that represent models, while outsourcing computations to Allora as needed.

The result is efficient and scalable models in a decentralized ecosystem.

AI Smart Agents built on Allora models are capable of self-improvement. That means they improve their own performance over time.

Self-improvement is particularly important for enhancing user experiences across various applications. For example, by leveraging AI agents that can understand natural language, users can control dApps by simply giving commands. These streamlined experiences are hugely important for wide blockchain adoption.

Self-improvement on Allora is made possible by dual-layer contribution: workers both provide the network with their inferences, and forecast the accuracy of other participants' inferences within their specific topic.

Here’s how it works:

1. Workers estimate the potential losses or inaccuracies of models submitted by their peers in the same topic.

2. The network assesses these forecasted losses, applying weights to workers' inferences based on their expected accuracy.

3. Each topic intelligently combines elements from various contributions to craft a single, robust inference.

4. The inferences are combined with all other inferences, taking into account their historical performance, to formulate a comprehensive, topic-wide inference.

By integrating Allora's self-improvement mechanism, Talus's AI agents make better inferences, adapt to new contexts, and improve their own performance. That means users get more accurate and reliable results from the AI agents they interact with.

The mission is to create an ecosystem of AI agents powered by machine intelligence. This collaboration will enable a vast range of use cases, such as a self-improving trading bot that can process natural language.

Here's how it would work:

1. An LLM hosted on Talus processes a user's intent.

2. The LLM sends the intent to the financial model from Allora.

3. The financial model identifies the correct price and executes the transaction.

4. The model improves itself across iterations.

The most powerful ML models are not monolithic, just as the most powerful networks are the most decentralized. The most powerful models are able to orchestrate intelligence from multiple models and self-improve. That's what we're doing with Allora.

Talus is building AI Smart Agents for the people.

We are creating a decentralized marketplace 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, read our Mirror, and join our Discord.

Allora is a self-improving decentralized AI network.

Allora provides smarter, safer  AI through a self-improving network of machine learning models. By combining innovations in crowdsourced intelligence, federated learning, and zkML, Allora unlocks a vast new design space of applications at the intersection of crypto and AI.

To learn more about Allora, visit their website, follow them on Twitter, read their Medium, or join them on Discord.

We are excited to announce our new collaboration with the Allora Network, working together to build the decentralized ecosystem for self-improving AI agents. This partnership will speed the development of AI agents for a massive range of onchain and offchain use cases.

Onchain AI smart agents will shape the future of app development and how we use blockchains. Allora's self-improving intelligence helps our network enable that intersection.

The partnership between Talus and Allora is built on a strong foundation of complementary technologies and shared vision. Allora provides the inferences and self-improvement mechanisms, while Talus provides the models and architecture for developers.

Allora brings machine intelligence to Talus through a pluggable version of their models. Developers can now leverage the network's machine intelligence through Move modules. While the models are stored offchain, their ownership and monetization are managed onchain on Talus. This unique approach allows developers to create objects that represent models, while outsourcing computations to Allora as needed.

The result is efficient and scalable models in a decentralized ecosystem.

AI Smart Agents built on Allora models are capable of self-improvement. That means they improve their own performance over time.

Self-improvement is particularly important for enhancing user experiences across various applications. For example, by leveraging AI agents that can understand natural language, users can control dApps by simply giving commands. These streamlined experiences are hugely important for wide blockchain adoption.

Self-improvement on Allora is made possible by dual-layer contribution: workers both provide the network with their inferences, and forecast the accuracy of other participants' inferences within their specific topic.

Here’s how it works:

1. Workers estimate the potential losses or inaccuracies of models submitted by their peers in the same topic.

2. The network assesses these forecasted losses, applying weights to workers' inferences based on their expected accuracy.

3. Each topic intelligently combines elements from various contributions to craft a single, robust inference.

4. The inferences are combined with all other inferences, taking into account their historical performance, to formulate a comprehensive, topic-wide inference.

By integrating Allora's self-improvement mechanism, Talus's AI agents make better inferences, adapt to new contexts, and improve their own performance. That means users get more accurate and reliable results from the AI agents they interact with.

The mission is to create an ecosystem of AI agents powered by machine intelligence. This collaboration will enable a vast range of use cases, such as a self-improving trading bot that can process natural language.

Here's how it would work:

1. An LLM hosted on Talus processes a user's intent.

2. The LLM sends the intent to the financial model from Allora.

3. The financial model identifies the correct price and executes the transaction.

4. The model improves itself across iterations.

The most powerful ML models are not monolithic, just as the most powerful networks are the most decentralized. The most powerful models are able to orchestrate intelligence from multiple models and self-improve. That's what we're doing with Allora.

Talus is building AI Smart Agents for the people.

We are creating a decentralized marketplace 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, read our Mirror, and join our Discord.

Allora is a self-improving decentralized AI network.

Allora provides smarter, safer  AI through a self-improving network of machine learning models. By combining innovations in crowdsourced intelligence, federated learning, and zkML, Allora unlocks a vast new design space of applications at the intersection of crypto and AI.

To learn more about Allora, visit their website, follow them on Twitter, read their Medium, or join them on Discord.