
相关推荐

Qlib
<p>Qlib 是一个开源的、面向 AI 的量化投资平台,旨在利用 AI 技术挖掘量化投资的潜力,赋能研究,创造价值,涵盖从探索想法到落地生产的全过程。Qlib 支持多种机器学习建模范式,包括监督学习、市场动态建模和强化学习。</p> <p>越来越多不同范式的 SOTA Quant 研究成果/论文正在 Qlib 中发布,以协作解决量化投资领域的关键挑战。例如,1)使用监督学习从丰富且异构的金融数据中挖掘市场复杂的非线性模式;2)使用自适应概念漂移技术建模金融市场的动态特性;3)使用强化学习建模持续投资决策并协助投资者优化交易策略。</p> <p>它包含完整的机器学习流程,涵盖数据处理、模型训练、回测等环节,并覆盖量化投资的整个流程:Alpha 寻求、风险建模、投资组合优化和订单执行。</p> <p>地址: <a href="https://github.com/microsoft/qlib" target="_blank" rel="noopener">https://github.com/microsoft/qlib</a> </p>

FinRobot
<p>FinRobot是一个超越 FinGPT 范畴的 AI 代理平台,是专为金融应用精心设计的综合解决方案。它集成了多种 AI 技术,超越了单纯的语言模型。这种广阔的视野凸显了平台的多功能性和适应性,能够满足金融行业的多方面需求。</p> <p>AI代理的概念:AI代理是一种智能体,它使用大型语言模型作为大脑来感知环境、做出决策并执行动作。与传统的人工智能不同,AI代理具有独立思考和利用工具逐步实现既定目标的能力。</p> <div class="markdown-heading" dir="auto"> <h3 class="heading-element" dir="auto" tabindex="-1">FinRobot 的整体框架分为四个不同的层,每个层旨在解决金融 AI 处理和应用的特定方面:</h3> <a id="user-content-the-overall-framework-of-finrobot-is-organized-into-four-distinct-layers-each-designed-to-address-specific-aspects-of-financial-ai-processing-and-application" class="anchor" href="https://github.com/AI4Finance-Foundation/FinRobot#the-overall-framework-of-finrobot-is-organized-into-four-distinct-layers-each-designed-to-address-specific-aspects-of-financial-ai-processing-and-application" aria-label="永久链接:FinRobot 的整体框架分为四个不同的层,每个层旨在解决金融 AI 处理和应用的特定方面:"></a></div> <ol dir="auto"> <li>金融AI代理层:金融AI代理层现已包含金融思维链(CoT)提示,增强了复杂的分析和决策能力。市场预测代理、文档分析代理和交易策略代理利用CoT将金融挑战分解为逻辑步骤,将其先进的算法和领域专业知识与不断变化的金融市场动态相结合,从而获得精准、可操作的洞察。</li> <li>金融 LLM 算法层:金融 LLM 算法层配置并利用针对特定领域和全球市场分析定制的专门调整模型。</li> <li>LLMOps 和 DataOps 层:LLMOps 层实施多源集成策略,利用一系列最先进的模型,为特定的财务任务选择最合适的 LLM。</li> <li>多源 LLM 基础模型层:该基础层支持各种通用和专用 LLM 的即插即用功能。</li> </ol> <div class="markdown-heading" dir="auto"> <h2 class="heading-element" dir="auto" tabindex="-1">FinRobot:代理工作流程</h2> <a id="user-content-finrobot-agent-workflow" class="anchor" href="https://github.com/AI4Finance-Foundation/FinRobot#finrobot-agent-workflow" aria-label="永久链接:FinRobot:代理工作流程"></a></div> <p> </p> <ol dir="auto"> <li> <p>感知:该模块捕获并解释来自市场信息、新闻和经济指标的多模式金融数据,使用复杂的技术来构建数据以进行彻底的分析。</p> </li> <li> <p>大脑:作为核心处理单元,该模块使用 LLM 感知来自感知模块的数据,并利用财务思路链 (CoT) 流程生成结构化指令。</p> </li> <li> <p>行动:该模块执行来自大脑模块的指令,运用工具将分析洞察转化为可操作的结果。行动包括交易、投资组合调整、生成报告或发送警报,从而积极影响金融环境。</p> </li> </ol> <div class="markdown-heading" dir="auto"> <h2 class="heading-element" dir="auto" tabindex="-1">FinRobot:智能调度器</h2> <a id="user-content-finrobot-smart-scheduler" class="anchor" href="https://github.com/AI4Finance-Foundation/FinRobot#finrobot-smart-scheduler" aria-label="永久链接:FinRobot:智能调度程序"></a></div> <p>智能调度器对于确保模型多样性以及优化每个任务最合适的 LLM 的集成和选择至关重要。</p> <ul dir="auto"> <li>主管代理:该组件协调任务分配过程,确保根据代理的绩效指标和特定任务的适用性将任务分配给代理。</li> <li>代理注册:管理注册并跟踪系统内代理的可用性,促进高效的任务分配流程。</li> <li>代理适配器:根据特定任务定制代理功能,增强其性能和在整个系统中的集成。</li> <li>任务管理器:管理和存储针对各种财务任务定制的不同通用和微调的基于 LLM 的代理,并定期更新以确保相关性和有效性。</li> </ul> <p>地址: <a href="https://github.com/AI4Finance-Foundation/FinRobot" target="_blank" rel="noopener">https://github.com/AI4Finance-Foundation/FinRobot</a> </p> <p>官方网址: <a href="https://ai4finance.org/" target="_blank" rel="noopener">https://ai4finance.org/</a> </p> <div class="markdown-heading" dir="auto"> <h2 class="heading-element" dir="auto" tabindex="-1">安装:</h2> <a id="user-content-installation" class="anchor" href="https://github.com/AI4Finance-Foundation/FinRobot#installation" aria-label="永久链接:安装:"></a></div> <p>1.(推荐)创建一个新的虚拟环境</p> <div class="highlight highlight-source-shell notranslate position-relative overflow-auto" dir="auto"> <pre>conda create --name finrobot python=3.10 conda activate finrobot</pre> <div class="zeroclipboard-container"> </div> </div> <p>2. 使用终端下载 FinRobot repo 或手动下载</p> <div class="highlight highlight-source-shell notranslate position-relative overflow-auto" dir="auto"> <pre>git clone https://github.com/AI4Finance-Foundation/FinRobot.git cd FinRobot</pre> <div class="zeroclipboard-container"> </div> </div> <p>3. 从源代码或 pypi 安装 finrobot 及其依赖项</p> <p>从 pypi 获取我们的最新版本</p> <div class="highlight highlight-source-shell notranslate position-relative overflow-auto" dir="auto"> <pre>pip install -U finrobot</pre> <div class="zeroclipboard-container"> </div> </div> <p>或者直接从这个 repo 安装</p> <div class="snippet-clipboard-content notranslate position-relative overflow-auto"> <pre class="notranslate"><code>pip install -e . </code></pre> <div class="zeroclipboard-container"> </div> </div> <p>4.修改OAI_CONFIG_LIST_sample文件</p> <div class="highlight highlight-source-shell notranslate position-relative overflow-auto" dir="auto"> <pre>1) rename OAI_CONFIG_LIST_sample to OAI_CONFIG_LIST 2) remove the four lines of comment within the OAI_CONFIG_LIST file 3) add your own openai api-key &lt;your OpenAI API key here&gt;</pre> <div class="zeroclipboard-container"> </div> </div> <p>5.修改config_api_keys_sample文件</p> <div class="highlight highlight-source-shell notranslate position-relative overflow-auto" dir="auto"> <pre>1) rename config_api_keys_sample to config_api_keys 2) remove the comment within the config_api_keys file 3) add your own finnhub-api "YOUR_FINNHUB_API_KEY" 4) add your own financialmodelingprep and sec-api keys "YOUR_FMP_API_KEY" and "YOUR_SEC_API_KEY" (for financial report generation)</pre> <div class="zeroclipboard-container"> </div> </div> <p>6. 开始浏览以下教程或演示:</p> <div class="snippet-clipboard-content notranslate position-relative overflow-auto"> <pre class="notranslate"><code># find these notebooks in tutorials 1) agent_annual_report.ipynb 2) agent_fingpt_forecaster.ipynb 3) agent_trade_strategist.ipynb 4) lmm_agent_mplfinance.ipynb 5) lmm_agent_opt_smacross.ipynb </code></pre> <div class="zeroclipboard-container"> </div> </div> <div class="markdown-heading" dir="auto"> </div>

✅ Official Alpha Vantage MCP Server
A MCP server for the stock market data API, Alphavantage API.

MCP Financial Advisor
A Model Context Protocol (MCP) server for financial advisory and bookkeeping, providing structured financial data and context for AI agents.

mcp-server-yahoo-finance MCP server
MCP Server for Yahoo Finance (STATUS: PRE-ALPHA)

Angle One MCP Server
Mirror of

Free USDC Transfer MCP Server
MCP (Model Context Protocol) server - free usdc transfer powered by Coinbase CDP

mcp-server-cex-bn
Mirror of

MCP Trader Server
A Model Context Protocol (MCP) server for stock traders

OptionsFlow MCP Server
An MCP server providing advanced options analysis through Yahoo Finance, supporting Greeks calculations, strategy evaluation (CCS/PCS/CSP/CC), and risk metrics. Built for MCP with Claude.ai.

mcp-server-cex-bn

Financial Analysis MCP Server
Mirror of

kospi-kosdaq-stock-server
An MCP server that provides KOSPI/KOSDAQ stock data using FastMCP

Alpaca MCP Server
Model Context Protocol (MCP) server for Alpaca trading API

Zerodha MCP Integration
Mcp server to connect with zerodha's kite trade apis

mcp-server-creem
Model Context Protocol (MCP) Server for Creem.io – Enhancing Merchant of Record services with transparent billing and improved fees.

mcp-server-ledger: A Ledger CLI MCP Server
A Model Context Protocol server for interacting with Ledger CLI, a powerful double-entry accounting system. This server enables Large Language Models to query and analyze financial data through a standardized interface, making it easy for AI assistants to help with financial reporting, budget analysis, and accounting tasks.
