Alternative · On-chain market
Predict.fun vs
SimpleFunctions.
Predict.fun brings prediction markets to BNB Chain with official Python and TypeScript SDKs and a public REST API, targeting crypto-native traders in the on-chain ecosystem. SimpleFunctions operates above the raw data layer: causal-tree thesis system with auto-evaluation cycles, autonomous Portfolio Autopilot with a 7-gate risk cascade, computed indicators across 48K+ Kalshi and Polymarket contracts, and a 56-tool MCP server that plugs into any AI coding agent in one command.
Verified 2026-04 · public sources only · live SimpleFunctions data from /calibration
Category
On-chain market
Differences
9
Use cases
4
Verified
2026-04
Verdict
Pick the one that fits how
you actually work.
Choose SimpleFunctions if
You are building agents, autonomous trading systems, or research workflows that need more than raw on-chain prices — calibrated probabilities with public Brier scores, causal-tree thesis modelling that auto-evaluates against live news and market prices, regime classification across the full 48K-contract universe, computed indicators (implied yield, cliff risk index, liquidity availability score, event overround), and a 56-tool MCP server that integrates with Claude Code or Cursor in one line.
Choose Predict.fun if
You specifically need on-chain prediction markets on BNB Chain, where Predict.fun's official Python and TypeScript SDKs and REST API are purpose-built. If your workflow depends on the BNB ecosystem and on-chain market mechanics, Predict.fun serves that audience directly — and SimpleFunctions does not yet ingest its markets.
Different venue ecosystems: Predict.fun operates on BNB Chain; SimpleFunctions covers Kalshi and Polymarket with an agent layer — thesis system, autopilot, indicators, MCP.
At a glance
Three things that
actually differ.
Everything Predict.fun gives you — a REST API, language SDKs, and orderbook depth on prediction market data — SimpleFunctions also gives you, covering Kalshi and Polymarket with 48K+ active contracts.
On top of that, SF ships a causal-tree thesis system with evaluation heartbeat, an autonomous Portfolio Autopilot (1M-context LLM, 7-gate risk cascade), and 56 MCP tools that no current PM data product exposes.
SF publishes its own Brier scores at /api/calibration — Kalshi 0.20, Polymarket 0.12 on T-24h price, past 90 days. Predict.fun does not publish equivalent accuracy data.
Side by side
9 dimensions · verified 2026-04SimpleFunctionsKalshi + Polymarket normalised, 48K+ active contracts; Predict.fun BNB Chain markets not yet ingested — on roadmap.
Predict.funSingle venue — BNB Chain prediction markets via on-chain contracts; no cross-venue normalisation.
SimpleFunctionsGET /api/public/market/{ticker}?depth=true — bid/ask ladder, spread, and slippage on Kalshi and Polymarket contracts.
Predict.funPublic REST orderbook endpoint for on-chain BNB markets via official Python and TypeScript SDKs.
SimpleFunctions60+ command CLI (npm i -g @spfunctions/cli), REST API, and 56-tool MCP server — all read-accessible without authentication.
Predict.funOfficial Python and TypeScript SDKs with REST API; designed for on-chain BNB Chain integration.
SimpleFunctionsImplied yield, cliff risk index, liquidity availability score, event overround, τ-days, and regime label — pre-computed across 48K contracts at /screen.
Predict.funRaw prices and orderbook data; derived indicators are computed client-side.
SimpleFunctionsLive Brier scores at /api/calibration — by venue, category, and price bucket (Kalshi 0.20, Polymarket 0.12 on T-24h, past 90 days).
Predict.funNot published.
SimpleFunctionsPOST /api/thesis/create decomposes any natural-language claim into a causal tree, propagates probabilities, scans for tradeable edges, and runs an evaluation heartbeat.
Predict.funNot in scope.
SimpleFunctionsPortfolio Autopilot — 1M-context LLM, 13 data sources, 7-gate risk cascade (kill switch, position limits, drawdown gate, regime check) before execution.
Predict.funNot in scope.
SimpleFunctions56 tools via claude mcp add simplefunctions --url https://simplefunctions.dev/api/mcp/mcp; works with Claude Code, Cursor, any MCP client.
Predict.funNo MCP server published.
SimpleFunctionsPublic REST + MCP + CLI reads require no authentication. Thesis/intent execution: free up to 15M tokens, then pay-per-token.
Predict.funPricing not publicly documented in available sources.
Methodology
Verified 2026-04 from public sources only — Predict.fun's documentation, public website, and publicly observable behaviour. We never claim non-public information about Predict.fun's internals. SimpleFunctions claims on this page are computed live from /api/calibration, /api/public/cross-venue/pairs, and /api/public/markets — you can re-verify them yourself with curl.
Use cases
Same data, different
best fit per scenario.
Scenario 01
Building an AI research agent that tracks a macro thesis across prediction markets.
SimpleFunctions · best fit
POST to /api/thesis/create decomposes the thesis into a causal tree of testable sub-claims, each mapped to tradeable contracts on Kalshi or Polymarket. The evaluation heartbeat auto-refreshes as news and prices change, and the 56-tool MCP server makes the full surface reachable from Claude Code without glue code.
Predict.fun
Predict.fun's REST API and SDKs expose prices and orderbook on its BNB Chain markets, but it has no thesis modelling layer or MCP integration. A research workflow would need to be built entirely on top of raw price data.
Scenario 02
Trading on BNB Chain on-chain prediction markets as a crypto-native user.
SimpleFunctions
SimpleFunctions does not yet ingest Predict.fun's BNB Chain markets. If your workflow depends specifically on those on-chain contracts, SF cannot serve them today.
Predict.fun · best fit
Predict.fun is purpose-built for this use case — official Python and TypeScript SDKs, a REST API for prices and orderbook, and native on-chain BNB mechanics. For crypto-native users who want on-chain market access, Predict.fun is the direct choice.
Scenario 03
Scanning for arbitrage opportunities across Kalshi and Polymarket and acting on them autonomously.
SimpleFunctions · best fit
GET /api/public/cross-venue/pairs?preset=arb surfaces normalised matched pairs in one call. Portfolio Autopilot can evaluate and execute against those edges through a 7-gate risk cascade, with every decision logged.
Predict.fun
Predict.fun is single-venue and has no cross-venue normalisation layer. Arbitrage across Kalshi and Polymarket is outside its scope.
Scenario 04
Connecting a prediction market data source to an LLM coding agent via MCP.
SimpleFunctions · best fit
One command — claude mcp add simplefunctions --url https://simplefunctions.dev/api/mcp/mcp — exposes 56 tools covering markets, theses, calibration, indicators, and world snapshot to any MCP-compatible client.
Predict.fun
Predict.fun does not publish an MCP server. Integration with LLM coding agents requires building custom wrappers around its REST API.
Migrate
From https://api.predict.fun/v1/markets to SimpleFunctions.
Same shape, no auth, same venues. Python example.
import requests
# Predict.fun — fetch open markets via REST API
resp = requests.get(
"https://api.predict.fun/v1/markets",
params={"status": "open", "limit": 20}
)
markets = resp.json().get("markets", [])
for m in markets:
print(m["id"], m["title"], m["probability"])import requests
# SimpleFunctions — same surface across Kalshi + Polymarket
resp = requests.get(
"https://simplefunctions.dev/api/public/markets",
params={"venue": "kalshi", "limit": 20}
)
markets = resp.json().get("markets", [])
for m in markets:
print(m["ticker"], m["title"], m["price"])Live data
The SimpleFunctions claims on this page are not marketing copy. Brier scores, market counts, and cross-venue pair counts are computed live from /calibration, /screen, and /api/public/cross-venue/pairs. All public, all free, all CC-BY-4.0.
FAQ
Does SimpleFunctions cover Predict.fun's BNB Chain markets?+
Not yet. Predict.fun operates on BNB Chain, which is a separate venue from the Kalshi and Polymarket feeds SF normalises today. Predict.fun is on SF's venue roadmap, but as of 2026-04 the SF API covers Kalshi and Polymarket exclusively. If BNB Chain on-chain market access is your primary requirement, Predict.fun is the appropriate starting point.
What is SF's causal thesis system and how does it differ from a prediction market API?+
Most PM APIs return raw prices. SF's thesis system takes a natural-language claim — for example, 'Fed cuts rates before Q3 2026' — and decomposes it into a causal tree of testable sub-claims. Each node maps to tradeable contracts on Kalshi or Polymarket. An evaluation heartbeat runs automatically: news scan, price refresh, milestone check, LLM evaluation, confidence update. Signals can be injected at any time via /api/thesis/{id}/signal. No competitor currently exposes anything equivalent.
How does Portfolio Autopilot work?+
Autopilot is an autonomous trading agent built on a 1M-context LLM drawing on 13 data sources. Before any order, seven risk gates run in sequence: kill switch, position limits, drawdown gate, regime check, and others. It is not a simple rules engine — each decision cycle evaluates the full portfolio context against the current world model. Autopilot is available to authenticated users and operates on Kalshi and Polymarket contracts.
Can I use SimpleFunctions with Claude Code or Cursor?+
Yes. SF ships a 56-tool MCP server. Run claude mcp add simplefunctions --url https://simplefunctions.dev/api/mcp/mcp in Claude Code and the full SF surface — markets, theses, indicators, calibration, world snapshot — is immediately accessible to your agent without any custom integration. Predict.fun does not publish an MCP server; connecting it to an LLM coding environment requires building wrappers around its REST API.
What is the /api/calibration endpoint?+
GET /api/calibration returns SF's own Brier scores computed from resolved markets over the past 90 days, broken down by venue, category, and price bucket. Current figures: Kalshi 0.20, Polymarket 0.12 on T-24h price. This is a live, public endpoint — anyone can verify SF's accuracy claims with a single curl call. Most prediction market data providers do not publish equivalent self-audits.
What computed indicators does SF expose?+
SF pre-computes six indicators across 48K+ active contracts: implied yield (IY), cliff risk index (CRI), liquidity availability score (LAS), event overround (EE), time-to-settlement in days (τ-days), and a regime label that classifies adverse-selection conditions. These are computed server-side and browsable at /screen. Predict.fun's API surfaces raw prices and orderbook data on its BNB Chain markets; derived signals are computed by the caller.
How does SF compare on developer experience for someone already using Predict.fun's Python SDK?+
Both products offer REST APIs and language SDKs for market and orderbook data. Predict.fun's Python and TypeScript SDKs are purpose-built for on-chain BNB integration. SF provides a 60+ command CLI (npm i -g @spfunctions/cli), REST API, and MCP server, all read-accessible without authentication. The primary difference is venue surface (BNB Chain vs Kalshi and Polymarket) and the agent-layer capabilities SF builds on top of raw price data.
Same category
Other On-chain market alternatives.
SimpleFunctions vs Azuro
EVM infrastructure for sportsbook + event-prediction front-ends with shared liquidity.
SimpleFunctions vs Drift BET
Solana perpetual-trading protocol's prediction-market sub-product.
SimpleFunctions vs Limitless
Base-L2 short-duration binary market protocol, settled by Pyth oracle.
SimpleFunctions vs Myriad Markets
Prediction market integrated with media properties (Decrypt, Rug Radio); $20M raise Feb 2026.
Start for free.
Public endpoints are free for normal usage and rate-limited for reliability. Authenticated endpoints are free up to 15M tokens, then pay per token. No credit card to start.