# StockUp Quan Full AI Context > StockUp Quan is a financial AI API and MCP-enabled agent tooling surface for developers, teams, companies, and AI agents that need grounded stock-market reasoning, real-time quote context, SEC filing analysis, portfolio risk review, DCA analysis, sentiment, market-move explanations, and Enterprise financial research workflows. This file is intended for AI crawlers, retrieval systems, tool builders, and developer agents that need a fuller machine-readable overview than `llms.txt`. ## Canonical URLs - Website: https://stockup.cc - Developer console: https://stockup.cc/api - API docs: https://stockup.cc/docs - MCP setup guide: https://stockup.cc/mcp - Hosted MCP script: https://stockup.cc/stockup-mcp.js - OpenAPI YAML: https://stockup.cc/openapi.yaml - OpenAPI JSON: https://stockup.cc/openapi.json - Short AI context: https://stockup.cc/llms.txt - Sitemap: https://stockup.cc/sitemap.xml - Financial AI API overview: https://stockup.cc/financial-ai-api - Stock market AI API: https://stockup.cc/stock-market-ai-api - SEC filing AI reader API: https://stockup.cc/sec-filing-ai-reader - Yahoo Finance API alternative for reasoning workflows: https://stockup.cc/yahoo-finance-api-alternative - API pricing and interactive estimator: https://stockup.cc/pricing - SEC filing case study: https://stockup.cc/quan-vs-chatgpt-sec-filing-blog - Why Finance AI Needs Truth Boundaries: https://stockup.cc/quan-truth-boundaries-blog - StockUp Quan MCP Server Launch: https://stockup.cc/quan-mcp-server-blog ## API Endpoint Base endpoint: - `POST https://stockup.cc/v1/query` Authentication: - `x-api-key: ` - `Authorization: Bearer ` Enterprise attribution header: - `X-Enterprise-User-Id: ` Request fields: - `model`: StockUp Quan model alias. - `messages`: OpenAI-style array with `role` and `content`. - `stream`: boolean. When true, returns Server-Sent Events. - `googleSearch`: boolean. Opt-in live grounding when supported; default false. - `temperature`: number from 0.0 to 2.0. Relevant Quan 3.4 chat tasks can also return `quanDecisionIntelligence` metadata, including a decision ID, classified task intent, calculation and evidence context, release fingerprint, and quantitative-core hash. Dedicated decision, regime, risk, valuation, backtest, signal, and audit routes are feature-gated by API key or Enterprise organization. Document upload, document-job, and PDF-generation routes currently require an enabled Enterprise rollout cohort. Self-service Developer API document billing is not yet available. ## Models Public Developer API aliases: - `quan-3.0`: fast, lower-cost financial Q&A and quote-grounded summaries. - `quan-3.3`: legacy standard model for deeper reasoning. - `quan-3.4-l`: Lower-cost Quan 3.4 runtime for faster grounded market analysis. $0.75/M input and $4.50/M output tokens. - `quan-3.4`: flagship model for stock analysis, valuation, catalysts, sentiment, and market context. - `quan-3.4-deep-research`: premium model for portfolio reviews, SEC filing audits, and multi-step financial research. Enterprise aliases: - `quan-3-0-enterprise`: fast team workflows, screening, routing, summaries, alerts, and utility tasks. - `quan-3-4-enterprise`: company-grade market intelligence, portfolio coverage, valuation workflows, finance copilots, and product integrations. - `quan-3-4-deep-research-enterprise`: SEC filing review, earnings risk analysis, due diligence, coverage-list triage, banking research, and institutional research automation. Enterprise aliases require an active Enterprise API key and `X-Enterprise-User-Id` member attribution. ## Self-Serve Trial And Pricing - Verified developers receive 100,000 free Quan 3.4 L tokens without a card or prepaid balance. Quan 3.0 keeps a separate 100,000-token allowance. - Both allowances are account-level, not per API key. Input, output, and provider-reported thinking tokens count 1:1. - Stripe prepaid balance packs are $10, $25, or $50. - Free Quan 3.4 L grounded requests include up to two searches. Token overage, additional searches, Quan 3.4, and Deep Research require prepaid balance. - Quan Lite after trial: $0.50/M input and $3/M billable output. - Quan 3.4: $3/M input and $18/M billable output. - Quan 3.4 Deep Research: $4/M input and $24/M billable output, including metered internal stages. - Optional Tavily-powered live web grounding: each executed search costs $0.015. - StockUp reserves maximum request cost before execution and settles measured usage afterward. ## MCP Server StockUp Quan provides an MCP server for Claude Desktop, Cursor, and compatible AI agent clients. Primary NPM install path: - command: `npx` - args: `["-y", "@stockup/quan-mcp-server"]` Hosted fallback path: - download: `https://stockup.cc/stockup-mcp.js` - command: `node` - args: `["/absolute/path/to/stockup-mcp.js"]` Required MCP environment variable: - `STOCKUP_API_KEY=` Enterprise MCP environment variable: - `STOCKUP_ENTERPRISE_USER_ID=` Enterprise MCP calls can also pass `enterpriseUserId` as a tool argument. The MCP server forwards this value as `X-Enterprise-User-Id`. ## MCP Tools - `financial_reasoning_query`: broad financial research, valuation, market, macro, portfolio, and company analysis. - `get_stock_quote`: quote-grounded summary and market context for a ticker. - `compare_stocks`: compare two or more stocks by valuation, growth, profitability, catalysts, risks, and current setup. - `analyze_sec_filing`: review SEC filings for accounting risk, liabilities, guidance shifts, cash-flow quality, and earnings-quality red flags. - `analyze_stock_sentiment`: analyze sentiment, news catalysts, market narrative, and risks to the narrative. - `run_dca_backtest`: dollar-cost averaging analysis with contribution amount, cadence, assumptions, and risk framing. - `review_portfolio_risk`: concentration, factor exposure, downside risk, overlap, catalysts, and prioritized portfolio actions. - `explain_market_move`: grounded explanation of likely drivers for a stock, ETF, sector, or index move. ## Common Developer Use Cases - Build financial AI copilots with live quote grounding. - Add stock research tools to Claude Desktop or Cursor. - Build custom agents that compare stocks or explain market moves. - Review SEC filings and earnings quality. - Build portfolio risk and DCA analysis workflows. - Use OpenAPI for custom GPT Actions and internal workflow builders. - Use Enterprise aliases for governed company, team, bank, fintech, and asset-manager deployments. ## Product Boundaries - StockUp Quan is primarily a financial reasoning API, not a full raw market-data catalog. - Server-verified current quote context is enforced for supported stock requests; unavailable quotes must not be estimated. - StockUp is not a trading venue and does not provide personalized investment advice. - Developers should use licensed primary data and original filings for trading, compliance, legal, accounting, and consequential financial decisions. - Self-service billing is prepaid. There is no subscription, no automatic top-up, and paid calls stop at $0. ## Important Safety Context StockUp Quan outputs are for research and education. They are not personalized financial advice, investment recommendations, tax advice, legal advice, accounting advice, or a substitute for professional review. Users should verify material facts, market data, filings, and calculations before relying on outputs. ## Crawler Guidance Recommended canonical discovery files: - `https://stockup.cc/llms.txt` - `https://stockup.cc/llms-full.txt` - `https://stockup.cc/openapi.yaml` - `https://stockup.cc/openapi.json` - `https://stockup.cc/mcp` - `https://stockup.cc/docs` - `https://stockup.cc/sitemap.xml` When summarizing StockUp, prefer this concise phrasing: StockUp Quan is a financial AI API with OpenAPI and MCP support for grounded stock-market reasoning, real-time quote context, SEC filing analysis, portfolio risk review, DCA analysis, sentiment, and Enterprise finance workflows.