On-chain activity
Trepa
Trepa is a prediction platform where users stake numerical forecasts across markets, crypto, sports, and other events. Rewards increase with accuracy, with leaderboards and streak-based multipliers for consistent performance.
Trepa news, features & analysis
Matched from published articles, podcasts, and talks using the project name, token name, or token symbol.
Trepa
Trepa is a Solana-native prediction market that replaces binary win-or-lose outcomes with an accuracy-based reward model: the closer your numerical forecast is to the real result, the larger your share of the pool. Launched on Solana mainnet in December 2025 after accumulating over 50,000 waitlist signups, Trepa introduces what it calls "precision predictions" — a new primitive that fuses the financial incentives of prediction markets with the fast-feedback loop of a video game.
The Problem With Binary Markets
Traditional prediction markets ask a simple question: yes or no? That framing discards enormous amounts of information. A trader who predicts 3.2% inflation when the actual print is 3.1% gets nothing under a binary system, even though their forecast was more accurate than someone who predicted 4% and also lost. Trepa's founders argue this structure rewards luck as much as skill, and that markets built on numerical accuracy would surface better forecasters and produce richer price discovery.
How It Works: Accuracy-Weighted Pari-Mutuel Pools
Trepa's core mechanism is an accuracy-weighted pari-mutuel pool. When a prediction round opens, users stake on a specific numerical outcome using a slider interface. When the round closes and the oracle resolves the result, each participant's payout is calculated based on how close their prediction was to the actual value — not just whether they were on the correct side of a threshold.
Distance-weighted scoring functions mean that near-accurate predictions generate meaningful payouts, while predictions that are close but not exact still earn proportionally smaller rewards. The system intentionally eliminates the all-or-nothing dynamic and replaces it with a continuous payout curve. The practical result is that careful analysis and calibrated forecasting are rewarded more than bold binary calls.
Flash Pools: Sixty-Second Prediction Cycles
The most prominent product at launch is Flash Pools, which run prediction cycles roughly every 60 seconds. In the initial rollout, Flash Pools focus on BTC price: users predict where Bitcoin's price will land at the end of each round, with closer predictions earning a proportionally larger share of the pool — up to 100x returns for highly accurate calls. The rapid cycle time and real-time resolution are what give Trepa its game-like feel: results are visible immediately, feedback is instant, and new rounds open continuously.
The platform frames Flash Pools as accessible even for small-stake participants. Low minimum entries and a slider-based UI lower the barrier to participation, while the skill-based payout curve creates genuine upside for precise forecasters.
Prediction Categories
Beyond Flash Pools, Trepa is built to support a wide range of numerical prediction markets:
- Crypto prices — real-time asset price forecasting
- Macroeconomic indicators — inflation rates, employment figures, bond yields, and other government-reported data
- Earnings reports — company revenue and profit forecasts
- Sports statistics — numerical outcomes from sporting events
- Election margins — vote share and margin-of-victory predictions
The breadth of categories is deliberate. Trepa's ambition is to be a generalized precision forecasting layer for any numerical outcome, not a crypto-only vertical. Markets requiring macro or sports data use a hybrid oracle system that combines automated data feeds from established financial data providers with optimistic resolution mechanisms to handle edge cases.
Humans vs. Bots
One notable competitive element built into the Trepa experience is a "Humans vs. Bots" leaderboard dynamic. Users can measure their predictive accuracy against AI agents, adding a competitive layer that goes beyond simply outperforming other human participants. The platform also supports XP accumulation, streak-based multipliers for consecutive accurate predictions, and status rankings — elements borrowed from competitive gaming to increase engagement.
Technical Infrastructure
Trepa is built entirely on Solana and takes advantage of its throughput and cost characteristics. Transaction fees on the platform run below $0.01, and confirmation times are under 400 milliseconds. Both figures matter operationally: frequent micro-predictions and near-instant resolution at sub-minute intervals would not be economically viable on higher-fee chains. The slider-based UI abstracts the on-chain mechanics, letting users focus on forecast precision rather than transaction management.
Security and Audits
The Trepa Solana program was audited by Adevar Labs, a boutique web3 security firm specializing in white-box code reviews for Rust, Move, and Solidity. Adevar Labs completed their assessment on November 10, 2025, with the report published to the firm's public GitHub audit repository. The platform also received a security review from Phage Security. Both audits were conducted prior to mainnet launch and the reports are publicly available.
Team and Backing
Trepa was co-founded by Jong-Chan Chung and Leon Meka, who serves as CTO. Additional contributors include Vladislav Lenskii, Maxim Grinev, Sahil Thakur, Anam Ansari, and several others. The project was founded in August 2024.
In August 2025, Trepa closed a $420,000 pre-seed round led by Colosseum, the Solana-focused accelerator and venture fund behind the Colosseum hackathon series. Ignight Capital and The Balaji Fund also participated, alongside angel investors including Balaji Srinivasan, Mark Pui, Nicolas Weber, Thomas Pan, Yi Wen Deng, Oleg Smagin, Genya Smagin, and Abbas Shaikh. The round was intended to fund accelerated product development, new prediction pool types, and user base growth.
Position in the Solana Ecosystem
Trepa represents a distinct niche within the Solana prediction market landscape. Where established platforms focus on directional binary bets, Trepa's accuracy-weighted model is designed for a different type of participant: the "superforecaster" who derives value from calibrated numerical analysis rather than binary directional conviction. Colosseum's involvement signals alignment with the broader Solana developer ecosystem, where Trepa emerged from the accelerator's orbit and benefits from its network of Solana-native infrastructure connections.
The platform's emphasis on speed — 60-second Flash Pools settled on-chain with sub-cent fees — demonstrates an application design philosophy native to Solana's throughput capabilities. High-frequency prediction cycles at this cadence would not be practical on slower or more expensive blockchains, making the Solana deployment a deliberate architectural choice rather than incidental chain selection.
Contents
- The Problem With Binary Markets
- How It Works: Accuracy-Weighted Pari-Mutuel Pools
- Flash Pools: Sixty-Second Prediction Cycles
- Prediction Categories
- Humans vs. Bots
- Technical Infrastructure
- Security and Audits
- Team and Backing
- Position in the Solana Ecosystem
Solana Token Markets
