FLOCK Officially Launches Web3-Optimized AI Model 'FLock Web3 Agent Model'

# FLock Launches Cutting-Edge AI Model for Optimized Web3 Operations FLock unveiled its large-scale AI model, the 'FLock Web3 Agent Model,' officially on the 12th, tailored for optimized Web3 operations. This model features precise function-calling capabilities, enabling it to effectively understand and execute complex Web3 tasks. The FLock Web3 Agent Model is trained within the FLock community's AI Arena. Users can participate as Task Creators, Training Nodes, Validators, and Delegates, collectively driving the model’s evolution. Built on Task1, contributors to the model receive rewards for their inputs. In the Web3 environment, unlike Web2, there is a shortage of computational resources, limiting the performance of both closed and open-source models. However, the FLock Web3 Agent Model overcomes these limitations and positions itself as a native Web3 AI expert. # Outperforming Competitors in Web3 AI Performance Metrics The FLock Web3 Agent Model registered a match rate of 75.93% in Web3 function-calling performance evaluations. This exceeds the performance of competitors such as GPT-4o, Gemini Flash 2.0, DeepSeek V3, and Qwen 2.5. These results demonstrate the model’s superior understanding, reasoning, and execution capabilities in Web3 tasks. Particularly, its ability to execute precise function calls ensures high accuracy. Both the model and evaluation criteria are open-source, guaranteeing transparency and accessibility. # Expanding the Web3 AI Ecosystem through Strategic Partnerships The FLock Web3 Agent Model is partnering with Io.net, OpenGradient, and HashKey Chain to accelerate the adoption of decentralized AI. This collaboration aims to broaden the scope of Web3-based AI applications. FLock intends to address structural issues in the AI industry using blockchain-based federated learning. The traditional AI industry faces challenges such as data acquisition difficulties, high entry barriers, and lack of transparency in model training. Federated learning allows individual devices to independently train AI models and share learned parameters, enhancing data privacy and model reliability. Despite the advantages, conventional federated learning relies heavily on central servers and faces challenges in securing trustworthy participants. FLock aims to solve these issues by applying blockchain technology. AI model proposals, training, validation, and utilization occur in a decentralized manner within the network, with participants rewarded based on their contributions. This approach seeks to democratize the AI industry and enhance AI model utilization across various sectors.