Skip to content
OpenAshare
ProductJournalOpen sourceTry the research desk
中文Try the research desk

Public beta

AI stock research for US, China and Hong Kong markets.

Ask about a stock or a headline. Bring price charts, available news, filings and cited sources into one research desk.

Try the research deskWatch the demo↓

MIT licensed. Research only, not investment advice.

Watch the demo↓

See how a question becomes research

Recorded in the research desk
00:00 | 00:30

See OpenAshare in 30 seconds.

Ask about a stock or headline, explore price charts, and follow the evidence in one research desk.

MIT / Open source

Read the code. Run it yourself.

OpenAshare is MIT licensed. Inspect the research workflow, adapt it, or run your own research desk with your own model key.

Read the source on GitHub ↗

Your own research desk

  1. Clone the repository
  2. Install Python and Node.js dependencies
  3. Configure your model key locally
  4. Start the API and web app
Follow the local setup guide ↗

Before your first question

Which markets does it cover?+

US, mainland China (A-shares) and Hong Kong equities. Price history is supported across these markets; company news and eligible US SEC filings depend on the symbol and provider. Support for a market does not mean every data type is available for every stock.

Where does the data come from?+

The code integrates Yahoo Finance via yfinance for US and Hong Kong prices, Finnhub for US news and fallback prices, AkShare and Chinese market sources for mainland data, and SEC EDGAR for eligible company filings. Availability depends on provider configuration. Prices may be delayed; this is not a real-time trading feed.

How do citations work?+

Research can include links to retrieved news, filings and other sources. Open the source and check its date, scope and reporting period. A citation is evidence to inspect, not a guarantee that an AI interpretation is correct. Missing evidence should remain a gap, not become a confident answer.

Is it free? What does public beta mean?+

The source code is free under the MIT License. Hosted AI research requires sign-in and available credits; check your balance in the research desk. Deep research is currently unavailable in the public beta. Self-hosted model and data providers may charge separately. We do not promise unlimited hosted use or future pricing.

Which AI model does it use?+

The model is configured by the deployment. The local setup documents DeepSeek and allows a model name, API base URL and API key to be configured. Hosted settings can change; a local example is not a guarantee of the hosted model. Compatibility with other providers depends on their API support.

What happens to my research data?+

The app supports local browser storage and account-backed workspace persistence when configured. AI requests send the question and relevant research context to the configured model provider. Self-hosting lets you choose that provider; it does not make external model or market-data requests stay on your device. Avoid entering confidential information.

Is this investment advice?+

No. OpenAshare is a research tool and does not place trades. AI answers can contain errors or omit important context. Verify original sources and make your own decisions.

Research journal

RESEARCH IDEA·BIOTECH / EVENTS

When a biotech headline changes the conversation

Explore the idea→
RESEARCH IDEA·TECH / SUPPLY CHAIN

Tracing the ripple effects of an AI chip cycle

Explore the idea→

Bring your next question.

Start with a company, a headline, or a period on a chart.

Try the research desk
OpenAshare
Data handlingFeedback on GitHub ↗GitHub ↗中文

For research and information only. Nothing here is a recommendation to buy, sell or hold a security. Market data can be delayed or incomplete, and AI-generated answers can be incorrect. Verify original sources. Investing involves risk, including loss of capital.

Data integrations: Yahoo Finance, Finnhub, AkShare and SEC EDGAR. Availability depends on configuration; no affiliation or endorsement implied.

Maintained by Zhiwei Chen ↗