简洁通用的群体智能引擎，预测万物
A Simple and Universal Swarm Intelligence Engine, Predicting Anything
MiroFish is a next-generation AI prediction engine for financial market forecasting, powered by multi-agent technology. By ingesting real market data (price history, macroeconomic indicators, research reports, policy statements), it automatically constructs a high-fidelity parallel market world. Within this space, thousands of intelligent agents — retail investors, institutional traders, fund managers, sell-side analysts, central bankers and financial media — with independent personalities, long-term memory, and trading logic interact and react to events. You can inject scenarios dynamically from a "God's-eye view" to deduce how prices, capital flows and market sentiment may evolve — rehearse the market in a digital sandbox before you make the decision.
You only need to: Pull live market data or upload seed materials (research reports, OHLCV CSV files, policy drafts) and describe your forecasting question in natural language
MiroFish will return: A detailed market prediction report and a deeply interactive high-fidelity market simulation
MiroFish is dedicated to creating a swarm intelligence mirror of financial markets. By capturing the collective emergence triggered by participant interactions, we break through the limitations of traditional forecasting:
- For Institutions: A risk-free rehearsal lab for stress-testing macro scenarios — rate decisions, earnings surprises, policy shifts — before committing capital
- For Individual Investors: An accessible sandbox to explore "what if" market scenarios across stocks, indices and the macro environment
From central bank decisions to single-stock catalysts, we let every market "what if" play out, making it possible to forecast the market.
Example scenarios: "If the Fed hikes 25bps at the next meeting, how will tech stocks and Treasury yields react?" · "If this company misses earnings guidance, how will institutional and retail flows diverge?"
Welcome to visit our online demo environment and experience a prediction simulation on trending public opinion events we've prepared for you: mirofish-live-demo
 |
 |
 |
 |
 |
 |
Click the image to watch the complete demo video for prediction using BettaFish-generated "Wuhan University Public Opinion Report"
Click the image to watch MiroFish's deep prediction of the lost ending based on hundreds of thousands of words from the first 80 chapters of "Dream of the Red Chamber"
Financial Prediction, Political News Prediction and more examples coming soon...
- Graph Building: Seed extraction & Individual/collective memory injection & GraphRAG construction
- Environment Setup: Entity relationship extraction & Persona generation & Agent configuration injection
- Simulation: Dual-platform parallel simulation & Auto-parse prediction requirements & Dynamic temporal memory updates
- Report Generation: ReportAgent with rich toolset for deep interaction with post-simulation environment
- Deep Interaction: Chat with any agent in the simulated world & Interact with ReportAgent
| Tool | Version | Description | Check Installation |
|---|---|---|---|
| Node.js | 18+ | Frontend runtime, includes npm | node -v |
| Python | ≥3.11, ≤3.12 | Backend runtime | python --version |
| uv | Latest | Python package manager | uv --version |
# Copy the example configuration file
cp .env.example .env
# Edit the .env file and fill in the required API keysRequired Environment Variables:
# LLM API Configuration (supports any LLM API with OpenAI SDK format)
# Recommended: Alibaba Qwen-plus model via Bailian Platform: https://bailian.console.aliyun.com/
# High consumption, try simulations with fewer than 40 rounds first
LLM_API_KEY=your_api_key
LLM_BASE_URL=https://dashscope.aliyuncs.com/compatible-mode/v1
LLM_MODEL_NAME=qwen-plus
# Zep Cloud Configuration
# Free monthly quota is sufficient for simple usage: https://app.getzep.com/
ZEP_API_KEY=your_zep_api_keyOptional — Market Data Sources:
# Stooq is used by default for stock/index/FX/commodity history — no API key required.
# Alpha Vantage (optional): free key at https://www.alphavantage.co/support/#api-key
ALPHA_VANTAGE_API_KEY=your_alpha_vantage_key
# FRED macroeconomic data (optional): free key at https://fred.stlouisfed.org/docs/api/api_key.html
FRED_API_KEY=your_fred_api_keyFetch live market data as simulation seed material via POST /api/graph/market-data/fetch, or upload an OHLCV .csv file directly — both are converted into market seed text automatically.
# One-click installation of all dependencies (root + frontend + backend)
npm run setup:allOr install step by step:
# Install Node dependencies (root + frontend)
npm run setup
# Install Python dependencies (backend, auto-creates virtual environment)
npm run setup:backend# Start both frontend and backend (run from project root)
npm run devService URLs:
- Frontend:
http://localhost:3000 - Backend API:
http://localhost:5001
Start Individually:
npm run backend # Start backend only
npm run frontend # Start frontend only# 1. Configure environment variables (same as source deployment)
cp .env.example .env
# 2. Pull image and start
docker compose up -dReads .env from root directory by default, maps ports 3000 (frontend) / 5001 (backend)
Mirror address for faster pulling is provided as comments in
docker-compose.yml, replace if needed.
The MiroFish team is recruiting full-time/internship positions. If you're interested in multi-agent simulation and LLM applications, feel free to send your resume to: mirofish@shanda.com
MiroFish has received strategic support and incubation from Shanda Group!
MiroFish's simulation engine is powered by OASIS (Open Agent Social Interaction Simulations), We sincerely thank the CAMEL-AI team for their open-source contributions!