Menu
farexio
Tech

Will Anthropic have the best AI model on the Chatbot Arena leaderboard at the end of July 2026?

Resolution: Updated:

In short

The market treats this as effectively settled in Anthropic's favor. With the check happening at noon ET on 31 July 2026 and the leaderboard already showing an Anthropic model on top, there is almost no time left for a rival lab to overtake it before the snapshot is taken.

Editorial illustration for: Will Anthropic have the best AI model on the Chatbot Arena leaderboard at the end of July 2026?

Probability

History starts collecting once the event is tracked

How the price has moved

The contract has traded at 100% since it was first recorded on 29 July 2026, with 435 price observations and a change of 0.0 percentage points over the most recent 24 hours. The only deviation in the full range was a brief touch down to 0%, which did not persist and is not explained by any single publicly reported event — it reads as a short-lived data or refresh artifact rather than a genuine reassessment. Aside from that one dip, the price history shows a market that considered this outcome close to certain from the start and has not found a reason to revise that view.

Analysis

Context

The Chatbot Arena leaderboard, run by LMArena, ranks large language models by pitting them against each other in blind, side-by-side conversations that ordinary users vote on. The resulting Arena score is one of the few public, continuously updated rankings of frontier AI models, and it gets cited constantly in coverage of the race between Anthropic, OpenAI, Google, xAI and Meta. This contract asks a narrow, mechanical question: whose model sits at the top of that specific leaderboard, with style control turned off, at a single moment — 12:00 PM ET on 31 July 2026. The leaderboard moves as labs release new models and as more votes accumulate, so the identity of the top model can and does change over weeks. Through July 2026 an Anthropic model has held the top position, which is why this contract has traded almost entirely at the extremes rather than settling into a genuine coin-flip range.
The clearest signal here is not the price level itself but its stability. The contract was first recorded on 29 July 2026 at 100%, and 435 separate price observations since then show almost no movement — the last 24 hours registered a change of exactly 0.0 percentage points. That kind of flat line at an extreme is what a market looks like when it considers a question functionally closed rather than genuinely contested; traders are not actively repricing new information because there is essentially none left to price before the 31 July check. The one notable feature of the data is the recorded range of 0% to 100% since the market opened. That means at some point the price touched zero — a brief window in which the market, or the underlying data feed, reflected something other than an Anthropic model on top, or the leaderboard was temporarily unavailable or unread. No single publicly reported trigger explains that dip, and it did not persist; the market returned to and has held at 100%. This is worth noting because it shows the mechanism can register sharp, short-lived swings around a leaderboard refresh, even when the settled outcome doesn't ultimately move. The structural reason confidence sits so high is timing. With settlement checked at a fixed moment — noon ET on 31 July 2026 — and only a day or so of runway left in July, a rival lab would need to both ship a new frontier model and have it accumulate enough Arena votes to overtake Anthropic's score before that exact snapshot. Chatbot Arena rankings shift with new model releases (GPT, Gemini and Grok updates have each briefly taken the top spot at various points over the past two years), but doing so requires a public release plus a data-gathering period; a same-day leapfrog this late in the month is a high bar. The single-venue volume of $1,204,750 on Polymarket, concentrated at the 100% extreme, indicates a market where liquidity has settled around consensus rather than around genuine disagreement. There is no second venue quoted here to check for a spread, so the entire visible market view is this one figure, and it points the same direction throughout.

What moves the probability

  1. Current leaderboard standing

    An Anthropic model already occupies the top rank on Chatbot Arena heading into the 31 July 2026 check. This is the dominant driver, since the contract asks only about the state at that one moment, not the whole month.

  2. Time remaining before the check

    With settlement fixed at noon ET on 31 July 2026, there is very little runway for a competing lab to release a new model and have it accumulate enough votes to overtake Anthropic's Arena score. This pushes the probability toward the extreme.

  3. Rival model releases from OpenAI, Google or xAI

    A surprise frontier release from a competitor in the final days of July could shift rankings quickly, since Chatbot Arena has seen leadership change hands before when a new model launches. This is the main scenario that could move the price away from its current level.

  4. Tie-break mechanics

    If two models finish with statistically close but technically distinct scores, ties are broken by the unrounded Arena score and then alphabetically by company name. This detail matters only in a near-tie scenario, which the current data does not suggest is close.

  5. Leaderboard availability at check time

    If lmarena.ai is inaccessible exactly at the 31 July 2026 check, resolution is delayed rather than defaulted, which affects timing but not the eventual outcome. This is a low-probability but non-zero operational risk.

The case for

  • An Anthropic model must still occupy the top rank on the Chatbot Arena leaderboard (style control off) when it is checked at 12:00 PM ET on 31 July 2026.
  • No rival lab ships and gains enough votes on a superior model in the remaining window before that check.
  • If scores end up close, Anthropic's unrounded Arena score must exceed rivals, or alphabetical tie-break rules must favor it.

The case against

  • A competing lab — OpenAI, Google or xAI — releases a new model in the final days of July 2026 that quickly accumulates votes and overtakes Anthropic's score.
  • A shift in the underlying vote data or a leaderboard recalculation before the check moment changes the ranked order in ways not yet visible.
  • If the leaderboard is unavailable at the exact check time, resolution is delayed, which does not change the outcome but extends uncertainty about when it is confirmed.

More about this event

Venues (1)

Probability

  • Will Anthropic have the best AI model at the end of July 2026?100%
  • Will OpenAI have the best AI model at the end of July 2026?0%
  • Will Google have the best AI model at the end of July 2026?0%
  • Will xAI have the best AI model at the end of July 2026?0%
  • Will Meta have the best AI model at the end of July 2026?0%
  • Will Moonshot have the best AI model at the end of July 2026?0%
  • Will DeepSeek have the best AI model at the end of July 2026?0%
  • Will Alibaba have the best AI model at the end of July 2026?0%
  • Will Microsoft have the best AI model at the end of July 2026?0%
  • Will Z.ai have the best AI model at the end of July 2026?0%
  • Will ByteDance have the best AI model at the end of July 2026?0%
  • Will Baidu have the best AI model at the end of July 2026?0%

What happened

Actual outcome: No

The market had this right

Resolution rules

Determined by
https://lmarena.ai/leaderboard/text
Resolution date

Resolution is based on the Chatbot Arena LLM Leaderboard at lmarena.ai/leaderboard/text, with style control off, checked at 12:00 PM ET on 31 July 2026. The contract resolves Yes if the top-ranked model at that moment belongs to Anthropic; ties are broken first by unrounded Arena score, then alphabetically by company name. If the leaderboard cannot be accessed at check time, resolution is delayed until it is reachable again.

Calculation methodology →

Local context

English-language tech and business media treat the Chatbot Arena leaderboard as a running scoreboard in the contest between Anthropic, OpenAI, Google, xAI and Meta, and headlines about which lab is

Common questions

What exactly settles this contract, and when?
The Chatbot Arena LLM Leaderboard on lmarena.ai, with style control turned off, checked at 12:00 PM ET on 31 July 2026. If the top-ranked model belongs to Anthropic at that moment, the contract resolves Yes.
What does a price of 100% actually mean here?
It means the market currently prices this outcome as essentially certain, based on where the leaderboard stands today and how little time remains before the check. It is not a guarantee — it is the collective view of everyone currently holding a position.
What happens if the leaderboard is down or unreachable at the check time?
The settlement rules specify that resolution is delayed until the leaderboard becomes accessible again, rather than defaulting to either outcome. This has not been needed so far.
Why did the price briefly show 0% if it has otherwise stayed at 100%?
The recorded range since the market opened spans 0% to 100%, meaning the price touched zero at some point, likely tied to a brief data or leaderboard refresh issue. It did not hold, and the price returned to 100% with no lasting effect on the trend.
How often does the Chatbot Arena leaderboard change its top-ranked model?
Historically, the top position has shifted whenever a lab ships a new frontier model that gathers enough user votes to overtake the incumbent score, which has happened periodically over the past two years among Anthropic, OpenAI, Google and xAI. With only about a day left before the July 2026 check, such a shift would need to happen very quickly to affect this outcome.
Can a position in this contract be closed before 31 July 2026?
Yes. Positions can generally be sold on the venue where they were opened at whatever price is quoted at the time, rather than held until the settlement check.

Related events