Will any AI model reach ___ Overall Arena Score by June 30?

Resolution Criteria

This market will resolve to "Yes" if any model on the Arena.AI Leaderboard (arena.ai/leaderboard/text) reaches at least the specified Arena Score by June 30, 2026, 11:59 PM ET. Otherwise, this market will resolve to "No". Results from the "Score" column under the "Text Arena | Overall" Leaderboard tab at https://lmarena.ai/leaderboard/text with style control off will be used to resolve this market. The resolution source for this market is the Chatbot Arena LLM Leaderboard found at arena.ai/leaderboard/text. If this resolution source is unavailable at check time, this market will remain open until the leaderboard comes back online and will resolve based on the first check after it becomes available. If permanently unavailable, this market will resolve to "No".

This market asks whether any AI model listed on the Chatbot Arena leaderboard (lmarena.ai/leaderboard/text) will reach specified Overall Arena Score thresholds — 1510, 1520, or 1530 — by 30 June 2026. Trading volume is most heavily concentrated on the 1510 threshold, with the higher thresholds attracting considerably thinner backing. Resolution is determined by the Score column under the Text Arena Overall tab with style control off.

Market structure

Three outcome tiers are tracked — scores of 1510, 1520, and 1530 — each resolving independently to Yes or No. Volume is heavily concentrated on the lowest threshold (1510), with the 1520 and 1530 tiers attracting broadly similar but much thinner interest. Resolution uses the official Chatbot Arena leaderboard at lmarena.ai/leaderboard/text, with style control disabled, checked at or after 30 June 2026 at 11:59 PM ET. If the leaderboard is temporarily unavailable, resolution is deferred until it returns.

Background

Chatbot Arena, operated by LMSYS and now accessible via arena.ai, is one of the most widely cited independent benchmarks for large language model capability. Its Elo-style scoring system — the Arena Score — is derived from millions of blind, head-to-head human preference votes rather than static test sets, making it resistant to deliberate overfitting. Scores have climbed steadily as frontier models from OpenAI, Google, Anthropic, Meta, and others compete for leaderboard position. The current upper tier of the leaderboard sits in a range that makes the 1510 threshold a live question, while 1520 and 1530 represent territory that no model had firmly occupied at the time these markets opened. The leaderboard updates continuously as new votes accumulate and new models are submitted.

Key factors

Several structural forces could influence whether any model crosses these thresholds before the deadline. First, the release cadence of frontier models from major AI laboratories is a key variable: a significant new release from any top-tier provider could shift the leaderboard rapidly. Second, Arena Scores are sensitive to vote volume — a model receiving a surge of comparisons can see its score stabilise or shift materially as statistical confidence grows. Third, the score thresholds are cumulative across all models, meaning a surprise entrant from a smaller or less prominent laboratory could trigger resolution. Fourth, the style-control-off condition means scores may differ from style-controlled figures, and models that benefit from stylistic presentation could be disadvantaged. Fifth, leaderboard methodology updates by the Arena team — including changes to the Elo calculation or eligibility criteria — could affect displayed scores. Finally, the June 30 deadline creates a fixed horizon; models released or significantly updated close to that date may not accumulate sufficient votes to move the score materially before resolution.

FAQ