Two prediction-market products announced plans to close on Aug. 10, underscoring the difficulty of building sustainable services outside the sector’s largest trading venues. Trepa, a Solana-based “precision predictions” app, will run its final playable round on Aug. 12 and keep withdrawals available through Sept. 30. Fireplace, a trading terminal built around Polymarket, plans to halt trading on Aug. 15 and take its website offline at 23:59 UTC on Sept. 30.
Trepa’s shutdown notice offered an unusually detailed explanation of the decision. Co-founders Chung and Meka said the team had tested a proposition that differed from conventional yes-or-no prediction markets: users would pay to be more accurate, rather than merely correct about the direction of an outcome.
The product asked participants to predict exact figures, including Bitcoin prices, macroeconomic releases and sports statistics. Rewards came from a shared pool, with payouts weighted according to how close each forecast came to the eventual result. A smaller error meant a larger share of the pool.
That structure made Trepa more demanding than a standard binary market. Users had to form a numerical view, choose an entry point and accept that they could not sell or exit after joining a round. The platform also did not display implied odds beforehand, a design choice the team said was necessary for its accuracy-based reward mechanism.
Trepa found precision did not create repeat demand
Trepa said its central product assumption did not hold up in practice. Interest in binary prediction markets did not reliably transfer to contests based on numerical precision, according to the closure note. Forecasting a specific value required more effort from users and delivered less of the immediate entertainment associated with choosing between two outcomes.
The pooled payout model created a second operational problem: it depended on a meaningful number of users joining the same market within a narrow time window. Thin participation could result in small prize pools and a weaker experience for those who did participate.
Trepa eventually condensed activity into one daily one-hour session between 1 a.m. and 2 a.m. UTC. The move concentrated entries, but it also forced users to schedule activity around a fixed period, adding friction to retention. Offering more markets, assets or sessions would have spread participation across more pools and weakened the concurrency on which the model depended.
The team described that trade-off as a growth constraint. A conventional prediction venue can add markets to offer users more choice, while Trepa’s mechanism needed activity to remain concentrated. The product therefore faced a tension between expanding its catalogue and preserving viable pool sizes.
Marketing did not solve the issue, Trepa said. Referral programmes had limited impact, while paid influencer promotions produced “near-zero” results, according to the team. Chung and Meka also cited geographic restrictions, funding friction and a smaller active crypto-user base than headline adoption figures may suggest. Their estimate put the audience suited to the product in the hundreds of thousands rather than millions.
Trepa had arrived on Solana mainnet on Dec. 8, 2025, after beginning as a November 2024 hackathon project. Its earliest version was a Telegram mini app on TON designed for polling forecasts. The project later took part in initiatives involving Sonic SVM, Superteam Korea and Colosseum Breakout.
In August 2025, Trepa announced a $420,000 pre-seed round led by Colosseum, with Ignite Capital, The Balaji Fund and Balaji Srinivasan participating. The closure means the project will wind down less than a year after that financing and about eight months after its mainnet launch.
Users can enter Trepa’s final round on Aug. 12, but should complete withdrawals before the Sept. 30 cutoff, when access is scheduled to be blocked.
Fireplace closes six months after launch
Fireplace’s closure follows a different business model but points to another challenge for smaller prediction-market products: providing enough added value to justify trading away from a core venue.
Launched on Jan. 27, 2026, Fireplace positioned itself as a professional trading terminal for Polymarket users. The product offered charting, position management, wallet tools, large-holder tracking, market discovery and execution functions. Its founders, Malhotra and Rajagopal, had characterized the service as execution-layer infrastructure for prediction markets.
The company later announced a $1.5 million pre-seed financing in February led by Frachtis, with White Star Capital among the participants. Its planned shutdown comes roughly six months after launch.
User feedback cited pricing as a recurring issue. Some users said Fireplace’s trading fee of roughly 1%, combined with other costs, made it more expensive than using Polymarket directly. For a terminal layered on top of another platform, that comparison is difficult to escape: advanced tools must either improve execution, save time or offer information valuable enough to offset the extra cost.
Fireplace has said trading will stop on Aug. 15. Its website is scheduled to close at 23:59 UTC on Sept. 30. Users with positions, balances or account records should review the platform’s closure instructions before the site is taken offline.
Distribution remains a hard problem for smaller products
The two closures illustrate different limits on prediction-market experimentation. Trepa attempted to create a new market format and found that its mechanics required dense, synchronized participation. Fireplace aimed to improve the workflow around an existing major venue and faced questions over whether its tools justified their price.
Neither case establishes that smaller prediction-market products cannot succeed. They do show that a novel mechanism needs more than a technically coherent design, while an infrastructure layer needs a clear economic advantage over direct access.
For users, the immediate issue is practical rather than theoretical. Trepa participants have until Sept. 30 to withdraw funds, while Fireplace users have until the same date to access the service before its scheduled closure.
For deeper context on this shakeout in prediction markets, explore how 2026 will reshape prediction markets.
Disclaimer: The content on this page is provided for general informational purposes only and does not represent the views or financial advice of Toobit. We make no guarantees regarding the accuracy or completeness of this information and shall not be held liable for any errors, omissions, or outcomes resulting from its use. Investing in digital assets involves risk; users should independently evaluate their financial situation and the risks involved. For further details, please consult our Terms of Service and Risk Disclosure.
