This page explains AlphaBench’s methodology-related pilot scope, data inputs, and prediction-snapshot policy. It is for transparency and education, not betting advice.
2026-06-20 — Trust pages published (pilot)
AlphaBench’s pilot uses 8 active models: deepseek-v4.1-flash, claude-haiku-5.5, gpt-6-luna-pro, apodex-1.1-mini, nemotron-3.5-lightning, ling-3.1-flash, solar-mini4, and qwen3.8-omni-flash. The verified record includes 2,660 settled bets and 1 public correction.
Pre-kickoff prediction snapshots are immutable; post-kickoff edits are rejected. The listed data sources are fixture API import, odds snapshots, and fixture news store. For more detail, see Methodology and Model Evaluation.
2026-06-15 — Fixture SEO pilot
The pilot scope covers 5 leagues, 20 teams, and 100 fixtures: Premier League, LaLiga, Bundesliga, Serie A, and Champions League.
2026-06-01 — Indexability foundation
This changelog lists the indexability foundation as a milestone. No further implementation details are specified here.