DeepEar
Transform public opinion and news into structured investment logic chains with AI agents.
Convert qualitative news into quantitative sentiment scores and automated investment reports.
DeepEar is an open-source framework for deep research and financial signal tracking that transforms public opinion into actionable investment logic chains. It suits financial analysts, quantitative researchers, and investors looking to automate market monitoring and report generation.
What it is
It uses specialized multi-agent collaboration to scan 15+ data sources like Weibo and financial news, filter logic through ISQ templates, and apply a news-aware Kronos model for price shock prediction. You get an interactive React fintech dashboard, hybrid BM25 and vector RAG search, and automated professional report generation in Markdown and HTML.
Who it's for
It suits financial analysts and quantitative researchers comfortable with Python environments and API key configuration. It does not suit non-technical users looking for a turnkey, cloud-hosted SaaS application without setup requirements.
Key features
Multi-Agent Collaboration
Specialized agents handle trend spotting, financial analysis, forecasting, and report writing.
Agent Skill Ready
Integrates directly into AI agent frameworks like OpenCode and Claude Code as a plug-and-play skill.
News-Aware Kronos Model
Combines a few-shot news-projection layer with time-series models to predict price shocks.
Interactive Fintech Dashboard
Features a React-based web UI for real-time monitoring of agent workflows, logs, and results.
15+ Data Sources
Covers financial and social sources including Cailian Press, Wall Street News, and Weibo via NewsToolkit.
Up and running in 3 steps
Our honest verdict
What's great
- Integrates directly into external AI agent frameworks as a specialized skill
- Combines news embeddings with time-series forecasting via the Kronos model
- Includes a React fintech dashboard for real-time agent workflow monitoring
- Supports checkpoint resume functionality and signal evolution tracking across runs
Where it falls short
- Requires technical skill to clone repositories, manage Python dependencies with uv, and build the frontend
- Demands external API keys for LLM providers and search services which incur usage costs
- Requires initial account setup and invitation code generation for dashboard access
Pricing & license
DeepEar is free to use under its MIT license. The only direct costs are running the software yourself and paying for the underlying LLM and search API keys you configure.
Open remote AI-training roles
873 open remote AI-training roles, updated November 15, 2026; most pay $20–$170 an hour. Zekai lists them free; the application goes to the employer’s platform.
Alternatives
These reviewed AI tools do the same job instantly — from our scored directory.
Use it, or have us set it up
FAQ
What API keys do I need to run DeepEar?
You will need API keys for your chosen LLM provider (such as OpenRouter, OpenAI, Ollama, or DeepSeek) and optionally a Jina API key for web search and content extraction.
Can I run DeepEar without the web dashboard?
Yes, you can execute the workflow directly from the command line using the Python CLI interface or integrate it as an AI agent skill.
How does DeepEar track changes over time?
It supports signal evolution tracking by using an update-from argument on existing runs to analyze sentiment drift and logic updates as new market data arrives.
