DSA - LLM for A/B Testing and Optimization

DSA - LLM for A/B Testing and Optimization

3.0

DSA uses LLM to automate A/B testing and optimization, enhancing decision-making.

Last Updated Nov 10, 2026
DSA - LLM for A/B Testing and Optimization screenshot

About

DSA (Deep Stock Analysis) is an intelligent stock analysis system powered by large language models (LLMs), designed specifically for investors in A-shares, Hong Kong stocks, and U.S. stocks. The system automatically integrates real-time market data, news sentiment, technical indicators, and fundamental data on a daily basis. Through AI reasoning, it generates individual stock decision references and market review reports, helping investors quickly grasp market trends amid vast amounts of information and improve decision-making efficiency.

Key Features

Multi-market coverage: Simultaneously supports major exchanges in A-shares (Shanghai, Shenzhen, Beijing), Hong Kong, and U.S. stocks without switching platforms.
Comprehensive data integration: Automatically collects stock quotes, financial data, technical indicators (such as MACD, RSI, moving averages), and news sentiment from across the web, presented in a unified view.
LLM-powered analysis: Uses large language models to cross-interpret multi-source data, generating daily bullish/bearish outlooks, risk alerts, and key support and resistance levels for individual stocks.
Daily decision reports: Outputs concise "decision cards" for each watched stock, including trend assessment, capital flow interpretation, and actionable recommendations.
Market review reports: Automatically generates daily recap reports for major A-share, Hong Kong, and U.S. indices, covering sector rotation, market sentiment, and key points to watch for the next trading day.

Pricing & Fees

  • Pricing: Not publicly listed on the pages we reviewed; check the official site for current plans.
  • Open-source deployment: DSA is distributed via GitHub and can be run free of charge through GitHub Actions, local execution, or Docker deployment.
  • Bring your own AI key: Users must configure their own AI provider key to generate reports, so LLM API usage costs are billed separately by the provider.
  • Data sources: Market data is pulled from free and community sources such as AkShare, Tushare, YFinance, TickFlow, and Longbridge, with fallback chains.
  • Notifications: Reports can be pushed to WeCom, Feishu, Telegram, Discord, Slack, or email at no extra platform fee.

Who It's For

Individual investors who want to save time on market review and quickly obtain multi-dimensional analysis references for stocks. Quantitative researchers can use LLM analysis output as supplementary input for fundamental and sentiment factors. Financial media creators can leverage DSA reports to quickly generate daily market commentary material. Investment advisors can use the system's output as an auxiliary analysis tool in client servicing.