Oblivia (our price prediction model) is a very very very experimental feature we rolled out, inspired by users’ requests to take advantage of our on-chain and social data to generate unified price predictions. today, we have 89% directional accuracy on Solana memecoin price predictions. predictions will be one of AI’s greatest strengths and you can’t rely on language models for it. that's why we combine LSTM models for short-term time series trends with forest-based classifiers. predictions run through a weighted ensemble calibrated for market regimes and liquidity shifts, and dynamic thresholding adapts to volatility, filtering low-conviction outputs in real-time. for more information on oblivia’s tech stack, check out predictions are validated on a monthly basis by @SaharaLabsAI, and we're also jointly building the SOL/USDT pair prediction stack with @AlloraNetwork. we’re currently working on a quant model to integrate directly into our predictions, therefore having a fully autonomous, real-time predictive execution stack. users can access our predictions page anytime and execute quick buy/sell directly. note: these are raw predictions, with no quant strategy or TP/SL logic embedded. predictions drop daily at 3:10pm CET, forecasting the next 24h price movement across selected tokens. if no tokens meet our precision threshold for a potential price increase, they won’t be displayed, so expect some days with fewer signals than others. once we have the quant strategy fully developed and merged with the prediction model, our predictions will become self-fulfilling, and we’ll be forced to only serve a small group of power users. regardless, there will be more spin-offs and products that will come out of this unified model. higher.