Qué es GFI

On September 9, Minara officially launched its next-generation universal financial agent, Minara Harness, which is now available for download by users worldwide. Minara Harness covers multiple asset markets and can conduct research and automated trading 24/7 within the scope authorized by users. Users can query financial markets using natural language, generate research reports, develop quantitative strategies, and handle daily tasks such as programming, web searching, and document management. The platform allows users to select their underlying large models. In terms of research, Minara Harness can simultaneously invoke multiple specialized AI agents to analyze the same investment opportunity from various perspectives, including fundamentals, valuation, market trends, and potential risks. Users can view the information sources and differing judgments of each agent and decide whether to execute trades based on the research results. Minara Harness also supports the writing of trading strategies in natural language, automatically generating code and completing backtesting. Users can subscribe to strategies from other traders, view strategy code, historical performance, and real-time operational metrics. Additionally, Minara offers an optional decision review mechanism that allows users to validate past judgments against subsequent market outcomes and assess which experiences can be applied to future research. Minara also released benchmark test results showing that Harness outperformed OpenClaw, Hermes, and Claude Code in several tests, demonstrating superior performance in tasks such as financial reporting, web research, and spreadsheets.

On September 9, Minara officially launched its next-generation universal financial agent, Minara Harness, which is now available for download to users worldwide. Minara Harness covers multiple asset markets and can conduct research and automated trading 24/7 within the scope authorized by users. Users can query financial markets in natural language, generate research reports, develop quantitative strategies, and handle daily tasks such as programming, web searches, and document management. The platform allows users to choose their underlying large models. In terms of research, Minara Harness can simultaneously call multiple specialized AI agents to analyze the same investment opportunity from various perspectives, including fundamentals, valuation, market trends, and potential risks. Users can view the information sources and differing judgments of each agent and decide whether to execute trades based on the research results. Minara Harness also supports the writing of trading strategies in natural language, automatically generating code and completing backtesting. Users can subscribe to other traders' strategies, view strategy codes, historical performance, and real-time operational metrics. Additionally, Minara offers an optional decision review mechanism that allows users to validate past judgments against subsequent market outcomes and assess which experiences can be applied to future research. Minara also released benchmark test results showing that Harness outperformed OpenClaw, Hermes, and Claude Code in multiple tests, demonstrating superior performance in tasks such as financial reporting, web research, and spreadsheets.

On September 9, CoinGecko, in collaboration with BlockBeats, released the RWA report titled "Tokenized Stocks: Insights from SpaceX, Pre-IPO Price Discovery, and the Surge of Perpetual Contracts." The report indicates that trading demand in the tokenized stock market is rapidly concentrating on perpetual contracts. In July 2026, the trading volume of stock perpetual contracts reached $376.3 billion, while the spot token trading volume was approximately $7.5 billion, making the former about 50 times larger than the latter; during the same period, the average open interest was around $6.1 billion. The report states that the monthly trading volume of stock perpetual contracts has increased approximately 209 times over the past 12 months, highlighting that leverage, short selling, cash settlement, and rapid onboarding capabilities are becoming the main reasons traders prefer such products. In contrast, spot tokens require 1:1 custody, minting, and redemption of the underlying stocks, leading to higher capital occupation and operational costs, while the onboarding speed of popular assets is also prone to limitations. The report further shows that on-chain stock perpetual contracts have become a 24-hour tool for trading industry themes. Trading in AI storage chip assets like SanDisk, Micron, and SK Hynix is active, reflecting that capital is chasing storage cycles, AI infrastructure investments, and macro market changes through on-chain perpetual contracts.