Which company has the best Code Arena WebDev AI model end of August?
Moonshot is the current leader, and the market has already priced it as the favorite, but the lead is not large enough to treat this as near-certain. I would still favor Moonshot, though slightly less strongly than the market does, because a late-August challenger from Anthropic or OpenAI could realistically overtake it.
Analysis
Moonshot enters this market as the incumbent leader on the arena.ai Code Arena WebDev leaderboard, with Kimi K3 reported to have taken first place in mid-July. The relevant market resolution depends only on who is ranked first at the August 31 check, so the key question is not who is best today but whether Moonshot can stay ahead through the rest of August. The current market price of 67% for Yes is consistent with that incumbent advantage, but it also implies traders see a meaningful chance of a late reversal.
The strongest case for Moonshot is that leaderboard leaders often keep their position long enough to resolve markets like this, especially when the lead is recently established and the model is already public. The reported gap to the nearest competitors is respectable but not overwhelming, which means Moonshot is not obviously vulnerable unless a rival launches a meaningful upgrade. If no major competitor release lands before the August 31 snapshot, Moonshot has a good chance of still being first simply by inertia and because the current top score is already competitive.
The strongest case against Moonshot is that the WebDev leaderboard has been highly dynamic, with Anthropic and OpenAI both appearing close enough to challenge quickly. The market horizon is only about five weeks away, which is long enough for one major model release to change the order but short enough that a single update can decide the outcome. Because the lead is not a runaway and because the field includes several strong labs with active release cadence, the probability of an upset is substantial.
Relative to the market, I am slightly more cautious on Moonshot because the current evidence supports it as a frontrunner but not a locked-in favorite. The most likely path is that Moonshot remains competitive and may still finish first, yet the combination of short time horizon, active rival labs, and leaderboard volatility makes a mid-60s probability more defensible than something higher. In other words, Yes is the likelier outcome, but the edge is moderate rather than dominant.
Arguments
For
- Arguments for Yes: Moonshot is already leading the relevant leaderboard, so it controls the base case going into the resolution date.
- Arguments for Yes: If competitors do not ship a decisive improvement in August, Moonshot can remain first through simple persistence.
Against
- Arguments against Yes: The leaderboard has active challengers close enough that a single update could change the ranking before August 31.
- Arguments against Yes: Recent model competition suggests the top spot can move quickly, making Moonshot vulnerable to a late surge from Anthropic or OpenAI.
Key drivers
- Moonshot currently holds the top WebDev position, which gives it the incumbent advantage.
- The remaining time until the August 31 check is short enough that one major rival release could still overturn the ranking.
Risk factors
- Anthropic or OpenAI could release a stronger WebDev model before month-end and reclaim first place.
- The current lead is meaningful but not large enough to rule out normal leaderboard volatility.
Scenarios
Best case
Moonshot keeps Kimi K3 at or near the top while rivals fail to produce a clearly superior WebDev release before the end-of-month snapshot.
Most likely
Moonshot remains one of the top contenders through August, but its final position depends on whether a major rival release lands before the resolution time.
Worst case
A competitor launches a materially better WebDev model in late July or August and overtakes Moonshot before the August 31 check.
More from this day
- politicsPolymarket10d
Iran leadership change by...?
AI86%MKT13%Edge+73Hidden GemThe market looks heavily tilted toward Yes because the supplied reporting already indicates that Iran’s top leadership has changed, with Mojtaba Khamenei named as supreme leader or, at minimum, the old leadership structure replaced by an interim council. The main uncertainty is definitional, but under the market’s broad language a Yes outcome is more likely than the current price suggests.
- PoliticsKalshi2y
Who will Trump pardon?
AI8%MKT51%Edge-43HypedI assess a low but nonzero chance that Barron Trump receives a presidential pardon before Jan. 21, 2029, because there is no known underlying criminal case that would make a pardon necessary. The market’s near-even price looks materially too high for a normal pardon outcome.
- PoliticsKalshi1y
2026: Trump's bad year?
AI58%MKT15%Edge+43Hidden GemI make the bear-case outcome moderately likely in 2026, at about 58%, because Trump is facing a growing stack of high-salience legal and institutional reversals rather than a single isolated controversy. The market’s 13% Yes price looks far too low if the event is defined as meaningful negative developments occurring during the year.