jangles-byte/Pythia

One local API call gives your agent the entire live state of the planet — every major world event at once, plus forecasts. 40+ free keyless feeds, runs entirely on Ollama. No cloud, no cost. Keyless live feeds into one global view and forecasts what’s likely next. No cloud, no keys, no cost. 1d,1w,1m,1y predictions by Mirofish.

What it solves

PYTHIA is a real-time global intelligence and forecasting system designed to solve the problem of information overload. While the world broadcasts vast amounts of signal (news, market movements, seismic data, etc.), no single person or system can monitor and reason across all these disparate streams simultaneously. PYTHIA fuses these feeds into a single "oracle" that predicts likely future events across multiple time horizons (24 hours to one year) with associated probabilities and reasoning.

How it works

PYTHIA integrates two open-source projects: Osiris (the "eyes"), which provides a 3D globe streaming over 40 live feeds, and MiroFish (the "mind"), a swarm-intelligence prediction engine. A local LLM (via Ollama) acts as the "voice" to interpret the world-state and generate forecasts.

Key technical components include:

  • Swarm Intelligence: A council of four specialist agents (Strategist, Economist, Naturalist, and Skeptic) deliberate on every forecast, providing consensus and highlighting dissent.
  • Local Execution: The entire system runs on the user's own hardware without cloud APIs or costs.
  • Brier-Weighted Consensus: The system tracks the accuracy of forecasts in a ledger and weights the votes of agents based on their historical performance.
  • Data Fusion: It ingests a wide array of keyless feeds, including SEC filings, federal contracts, market odds (Polymarket/Kalshi), and environmental sensors.

Who it’s for

  • AI Agent Developers: Who can use PYTHIA's machine-readable API to give their agents real-world context and "eyes on the planet."
  • Analysts and Strategists: Who need a fused, real-time view of geopolitical, economic, and environmental risks.
  • Power Users: Who want a local, private intelligence dashboard for monitoring global events and their potential knock-on effects.

Highlights

  • Multi-Horizon Forecasting: Predicts events across 24h, week, month, and year horizons with probability drift charts.
  • Live Global Dashboard: A 3D globe with map overlays, 3D altitude for satellites/aircraft, and access to 4,000+ public cameras.
  • Real-time Signal Rules: Customizable notifications for specific triggers (e.g., earthquake magnitude or market volatility).
  • Hypothetical Scenarios: A "what if" feature that allows users to inject scenarios into the live world to forecast counterfactual effects.
  • Comprehensive Feed Integration: Fuses 48+ free feeds including CISA KEV, NOAA space weather, and SEC insider trades.

Related

  • Project
    koala73/worldmonitor

    An AI-powered real-time global intelligence dashboard that aggregates news, geopolitical monitoring, and infrastructure tracking into a unified situational awareness interface.

  • Project
    666ghj/MiroFish

    A multi-agent prediction engine that creates high-fidelity digital sandboxes to simulate and predict the outcomes of real-world events or fictional scenarios.

  • Project
    666ghj/BettaFish

    BettaFish is an open‑source, Python‑based multi‑agent platform that automatically crawls social‑media (Weibo, TikTok, etc.), performs multimodal and sentiment analysis with LLMs and fine‑tuned models, lets agents collaborate through a forum‑style LLM moderator, and generates interactive HTML/PDF reports. It ships with Docker, Streamlit demos, and a modular architecture (QueryEngine, MediaEngine, InsightEngine, ReportEngine, ForumEngine, MindSpider crawler).

  • Project
    VRSEN/OpenSwarm

    A multi-agent system that coordinates specialized AI agents to generate professional deliverables like slide decks, research reports, and videos from a single terminal prompt.

  • Project
    kyegomez/swarms

    A production-ready multi-agent orchestration framework that provides pre-built architectures for coordinating multiple AI agents in sequential, concurrent, or hierarchical workflows.