CryptoWorld.net January 5 News: As 2026 approaches, the AI security field welcomes its first heavyweight benchmark test. LISABench announced today the launch of the Q1 2026 evaluation, aiming to identify the top-performing cutting-edge AI models in Web3 smart contract vulnerability detection through practical testing. Additionally, LISABench has launched a community voting reward activity for prediction forecasts. The evaluation lineup is luxurious, gathering KIMI K2 (Moonshot AI), DeepSeek V3.2 (Deep Exploration), QWen 3 30b-a3b (Alibaba Cloud), GLM 4.6 (Zhipu AI), GPT-5.2 (OpenAI), Gemini-3-pro-preview (Google), and Claude 4.5 Sonnet (Anthropic), the top 7 frontier models worldwide competing on the same stage. Currently, the prediction voting channel for the Q1 winners has been opened. Meanwhile, LISABench’s evaluation standard codebase has been open-sourced on GitHub for developers to review and reproduce.
View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
LISABench 2026 First Test of the Year, Seven Top AI Models Compete for the Web3 Vulnerability Detection King
CryptoWorld.net January 5 News: As 2026 approaches, the AI security field welcomes its first heavyweight benchmark test. LISABench announced today the launch of the Q1 2026 evaluation, aiming to identify the top-performing cutting-edge AI models in Web3 smart contract vulnerability detection through practical testing. Additionally, LISABench has launched a community voting reward activity for prediction forecasts. The evaluation lineup is luxurious, gathering KIMI K2 (Moonshot AI), DeepSeek V3.2 (Deep Exploration), QWen 3 30b-a3b (Alibaba Cloud), GLM 4.6 (Zhipu AI), GPT-5.2 (OpenAI), Gemini-3-pro-preview (Google), and Claude 4.5 Sonnet (Anthropic), the top 7 frontier models worldwide competing on the same stage. Currently, the prediction voting channel for the Q1 winners has been opened. Meanwhile, LISABench’s evaluation standard codebase has been open-sourced on GitHub for developers to review and reproduce.