Fetch.ai, ASI Alliance launch web3-native LLM designed for agentic AI

Fetch.ai, a founding member of the Artificial Superintelligence Alliance, has launched what it describes as the world's first web3-native large language model (LLM), the ASI-1 Mini.

The ASI Alliance is a collective formed by Fetch.ai alongside SingularityNET, Ocean Protocol and CUDOS. Their respective tokens were temporarily consolidated into Fetch.ai's FET token last year as a preparatory step for the final transition to a unified ASI token.

ASI-1 Mini is specifically built and optimized for supporting complex agentic workflows, designed to enable the web3 community to invest in, train and own advanced AI models in a decentralized way, according to a statement shared with The Block.

"ASI-1 Mini is the first major product from the ASI Alliance's innovation stack, marking the beginning of the ASI:<Train/> rollout and a new era of community-owned AI,” Fetch.ai CEO and Chairman of the ASI Alliance Humayun Sheikh said. "This launch sets the foundation for a decentralized ecosystem where the web3 community can invest, train and directly benefit from cutting-edge AI models."

Matching the performance of top LLMs

The project claims that benchmarking results show that ASI-1 Mini matches the performance of the top LLMs on the market while offering significantly lower hardware costs and web3 integration that allows users to interact with AI securely and autonomously.

Powered by the FET token via an integration with the ASI wallet, ASI-1 Mini aims to enable users to benefit from the growth of such models that could "unlock multi-billion-dollar valuations, creating a new era of decentralized ownership and shared value in AI technology," the project said.

Features include built-in support for four dynamic reasoning modes: Multi-Step, Complete, Optimized and Short Reasoning — automatically selecting the best approach based on the task. ASI-1 Mini is available from Tuesday as a tiered freemium product for FET holders, with additional functionality expected to roll out over the coming weeks.

Additionally, ASI-1 Mini will also be integrated with AgentVerse, Fetch.ai's agent marketplace, allowing users to create and deploy autonomous agents that execute real-world tasks from simple language commands. Once live, users can request tasks like booking reservations, which will be handled by micro-agents within AgentVerse. Developers can also monetize these agents, fostering an open-source ecosystem with accessible, customized AI functionality, according to the team.

"ASI-1 Mini is just the start — over the coming days, we will be rolling out advanced agentic tool-calling, expanded multi-modal capabilities and deeper web3 integrations," Sheikh added. "With these enhancements, ASI-1 Mini will drive agentic automation while ensuring that AI’s value creation remains in the hands of those who fuel its growth."