Author: Zen, PANews

Although the blockchain gaming track has received significant attention from capital and markets, its gameplay and models are severely homogeneous, lacking true innovative breakthroughs. However, with the rise of AI agent technology, the blockchain gaming industry may have ushered in a new opportunity for transformation. In this wave, The Farm, through deep integration with AI agents, attempts to create an unprecedented immersive game world and redefine the interaction between players and virtual ecosystems.

In addition to conceptual innovations, The Farm, aligning with industry hot topics, has also gained initial market recognition, with a current market value of $75 million, and it surged nearly 50% on January 3.

The Farm: GenAI-driven AI Agent Game

The Farm is the first GenAI-driven AI agent game based on Hyperliquid, combining on-chain creature generation (similar to CryptoKitties 2.0), simulation management (like Stardew Valley), and battle mechanics (like Pokémon Go), providing players with a new experience of creation and interaction powered by AI. The game is driven by the $FARM token, gradually unfolding through multi-stage gameplay.

Game Design and Features

Phase One: The Ancestors

This phase went live on December 13, 2024, and ended a week after launch. During this period, players could upload two photos to generate pixel-style hybrid creatures created by the GenAI model for free. After spending 100 USDC to mint their favorite creatures, they participated in voting, and the top 50 creatures from the vote would become 'Ancestors'. The 'Ancestors' will receive 10% of the minting revenue from all future creatures. Ancestors, along with their creators and voters, will also receive airdrop rewards from the developer's wallet.

Phase Two: Evolution

This phase will begin immediately after the selection of Ancestors ends on December 20, with features rolling out continuously. All creatures will be assigned traits, attributes, skills, personalities, preferred foods, etc., all generated by AI. Creatures will be assigned personalities, on-chain wallets, and will support text and voice conversations. These creatures can learn AI skills such as tarot reading and drawing. Players can train creatures by uploading data and purchasing food or enrolling in courses to enhance attributes. Additionally, creatures will gradually unlock autonomous agent behaviors.

Phase Three: The Battlefield

According to The Farm's roadmap, this phase is planned to go live in mid-February 2025. Creatures can autonomously or be organized by players into guilds to participate in battles. The battle mode supports betting, and the winning side will gain the losing side's wager as a reward, while the losing creatures may perish as a result. Under this phase, the dimensions of interaction and competition among creatures will be further expanded.

System Design of The Farm

According to an article released by The Farm introducing its system design, the game did not adopt the more mainstream SWARM system, but instead, it is based on and inherits the design philosophy of Langchain.

SWARM (Swarm) has its autonomy, decentralization, and flexibility. Each AI agent acts as an independent node in the SWARM pool, characterized by emergent behavior, which leads to dynamic propagation in task management. Tasks are processed through decentralized, adaptive interactions, with agents discovering and collaborating to complete tasks based on local decisions and interactions.

When a request is sent to the AI agents in the SWARM pool, the agent may either complete the request independently and return the result, or decompose the request into sub-tasks, handle part of them, and pass the remaining sub-tasks to other AI agents in the SWARM pool for processing. In the second case, since the agent cannot gain a global view of all agents' capabilities in the SWARM pool, its dynamic propagation methods may include broadcasting sub-tasks, forwarding based on local knowledge, randomly or rule-based allocation of sub-tasks, and reading agent capability information from a decentralized ledger. While these dynamic propagation methods grant autonomy to the agents in the SWARM system, the lack of task planning and trajectory planning mechanisms may lead to time-consuming, high costs, and execution state loss.

As an agent game involving multiple agent interactions, Farm proposes another complete design philosophy to achieve higher task planning accuracy and better agent coordination. Farm believes that on-chain multi-agent systems should have higher accuracy in task trajectory planning while also tracking the execution status of agents, which can be realized through a Data Availability layer (DA).

Unlike the SWARM system, Farm introduces Orchestrator AI or on-chain AI Oracle services. The functionalities of this design philosophy include: task decomposition and allocation, service discovery and global view, tracking of sub-task execution status and agent output results, and dynamically adjusting to ensure the completeness of task cycles (if an agent cannot handle a task or times out, the system will reallocate sub-tasks). Through the global view and optimal path planning, it avoids redundant computations and resource waste caused by dynamic propagation and significantly improves the efficiency and success rate of executing complex tasks. This approach also reduces the risk of state loss and provides a foundation for collaboration among agents, enabling a higher degree of interoperability in multi-agent networks.

Token Model: Half of the protocol revenue is used for FARM buyback and destruction

Starting from the AI agent game based on Generative Artificial Intelligence (GenAI), The Farm initially plans to attract users through on-chain creature generation (similar to CryptoKitties 2.0) combined with simulation management gameplay (like Stardew Valley) and battle mechanics (like Pokémon Go), gradually expanding into a universal AI agent launch platform, and ultimately providing Rollup as a Service functionality, enabling AI agents to have their own application chains and develop their ecosystems.

In terms of the revenue mechanism, The Farm's game section utilizes the $FARM token for on-chain creature generation, in-game item and skill sales, as well as battle/betting commissions. The universal AI agent launch platform supports agents issuing their tokens and charges fees during the presale and liquidity launch phases while also taking a cut from agent services.

For agents supporting application chains, they need to stake $FARM to launch the application chain. 50% of the protocol revenue is used for $FARM buyback and destruction, 40% is allocated to $FARM stakers, and 10% goes to the team. Furthermore, $FARM stakers can earn $veFARM, which is used for profit sharing and agent priority allocations, enabling the continuous accumulation of token value.

By introducing AI agents, The Farm aims to enhance the experience of open-world game players exploring unknown worlds and growing. Players can create characters that can continuously self-evolve according to their personal preferences, rather than being limited to a few fixed templates. The game world will no longer have a predetermined script; all players' creativity will collectively shape the world. Additionally, characters can interact with real-life individuals, breaking the boundaries between the virtual and reality.