Open-Sourcing BitQuant – Your Personal Quant Agent

Today, we’re open-sourcing BitQuant — the AI-native quant framework that turns natural language like: 

• “optimize my portfolio”
• “hedge my SOL exposure”
• “deposit 20k into Kamino”

into verifiable, onchain outputs.

After months of beta testing with 50k+ users, the full stack is now live under an MIT license: agents, prompt templates, protocol connectors — all of it. You can find the GitHub repository here. If you wish to interact with the agent first, you can do so here.

Why BitQuant Exists

DeFi is fast. Fragmented. Unforgiving.
Dashboards lie, bots break, risk hides in complexity.

BitQuant brings structure:

‍

Under The Hood

Decision Layer
Router LLM routes prompts to the right specialist agent:

• Analytics Agent — provides market-trend analysis, actionable intelligence, and risk insights
• Investment Agent — contains actionable trade logic (Solana first, EVM next)

Data & Execution Layer

• Native data & liquidity: Solana RPC, Orca, Solend, Kamino, CoinGecko, DeFiLlama
• API endpoint: POST /agent/run (FastAPI) — connect dashboards, wallets, or bots
• Decentralized inference: Subnet-ready via Bittensor hooks

Get Started in 60 seconds

git clone https://github.com/OpenGradientAI/bitquant
make venv && source venv/bin/activate
make install && make run       
make sample                    
make test                      

What You Can Build

BitQuant supports a wide range of AI-native DeFi agents out of the box:

• DeFi Copilots — prompt-based portfolio management that executes on your behalf
• Onchain Risk Dashboards — ask “What’s my liquidation risk?” and get an actionable answer
• Automated Yield Agents — extend the Investment Agent with new chains, pools, and strategies
• Research Bots — surface funding-rate skew, momentum decay, liquidity crunch alerts in real time

Core agent types include:

• Analytics Agent: Track TVL, yields, protocol risk, and portfolio volatility across chains. Example: “Analyze my portfolio’s risk profile.” / “Show TVL trends for Solana DeFi.”
• Investment Agent: Identify yield opportunities, compare protocols, and guide users through execution.
  Example: “Which pools offer the best stablecoin yield?” / “Compare Kamino and Orca for USDC/SOL.”
• Custom Agents: Build your own for trading, alerts, NFT analytics, cross-chain views, or entirely new workflows.

Agents are modular and composable — route them, extend them, remix them.

We’d love to hear from you if you build on BitQuant. We support contributors with technical feedback, grants, investor intros, and more.

Roadmap & Contribution

PRs are welcome — BitQuant is a canvas. Bring your tools! 

• Code: https://github.com/OpenGradient/BitQuant
• Updates: x.com/BitQuantAI
• Docs: /docs in the repo — help us expand them
• Use the platform: https://www.bitquant.io
• OpenGradient Discord: https://discord.com/invite/2t5sx5BCpB

We are looking forward to hearing from you!