August 2026 Best AI Model
Anthropic has maintained its lead on the AI model leaderboard despite a wave of new frontier releases over the past month. Claude Fable 5 still tops the Artificial Analysis Intelligence Index, while OpenAI narrowed the gap with GPT-5.6 and Moonshot AI entered the race with Kimi K3. xAI also shipped Grok 4.5, leaving six AI labs with models scoring above 50 on the benchmark for the first time. Polymarket now prices Anthropic at 100.0% to hold the best LLM spot through April 30, with $61.0K in cumulative volume on the April market. Our cross-platform odds for this month's "Best AI Model by Company" market aggregates live probability blended from Kalshi and Polymarket. DeFi Rate LLM prediction data is updated every 30 minutes.
Top prediction markets with AI model trading (daily, weekly, monthly)
Trade $10, Get $15
T&C’s apply.
Deposit $10 get a $50 trading bonus
Risk Disclaimer
Current Odds Snapshot
Current probabilities across platforms with liquidity indicators
CO5claude-opus-5-max
Vol $13.2K
Spread —
Agg
100.0%Settled
K
100.0%
CF5claude-fable-5
Vol $14.7K
Spread —
Agg
0.0%Settled
K
0.0%
CO4claude-opus-4-6-thinking
Vol $8.2K
Spread —
Agg
0.0%Settled
K
0.0%
CO4claude-opus-4-7
Vol $12.2K
Spread —
Agg
0.0%Settled
K
0.0%
CO5claude-opus-5-high
Vol $12.7K
Spread —
Agg
0.0%Settled
K
0.0%
| Outcome | Aggregated | Spread | Volume | Kalshi |
|---|---|---|---|---|
CO5 claude-opus-5-max | 100.0% Settled | — | $13.2K |
Kalshi
100.0% |
CF5 claude-fable-5 | 0.0% Settled | — | $14.7K |
Kalshi
0.0% |
CO4 claude-opus-4-6-thinking | 0.0% Settled | — | $8.2K |
Kalshi
0.0% |
CO4 claude-opus-4-7 | 0.0% Settled | — | $12.2K |
Kalshi
0.0% |
CO5 claude-opus-5-high | 0.0% Settled | — | $12.7K |
Kalshi
0.0% |
Probability Over Time
Hover for details · Cursor-synced tooltips
Tap "Chart settings" to adjust the chart.
Chart settings
Anthropic extends its streak in the #1 AI model market
Kalshi and Polymarket both run monthly markets on which company holds the top-ranked AI model on the Arena leaderboard. Anthropic has taken every month since February, and is on track to extend that streak again when the July market resolves on July 31.
Claude Opus 4.8, which launched in late May, and Claude Fable 5, which followed on June 9, have kept Anthropic’s models at the top of the Arena leaderboard through the summer — even through a bumpy stretch that included a brief U.S. export-control suspension of Fable 5 access in mid-June and a wave of “AI kill switch” legislative headlines. Neither event moved the leaderboard. On Polymarket, Anthropic has traded in the 80s-to-high-90s% range to hold the top spot through July 31, with Google and OpenAI splitting most of the remainder in the low single digits.
Cumulative Polymarket volume on the July market is running roughly $8M so far, continuing the steady pullback from the $20M+ months traders saw over the winter, when the race between Anthropic and Google was still competitive.
One difference worth noting: Kalshi resolves to the specific model (e.g., claude-opus-4-8, gemini-3.1-pro-preview), while Polymarket resolves to the company (Anthropic, Google). Kalshi also runs both variants — a model-level market and a company-level market. Our charts are aggregating data from the model market.
Monthly AI prediction results
| Month | Kalshi | Polymarket | Notes |
|---|---|---|---|
| July 2026 | claude-opus-5 ($898K) | Anthropic (100%, $8.17M) | Claude Opus 5 launched July 24 and immediately topped Artificial Analysis’s rebased v4.1 leaderboard, arriving just before month-end |
| June 2026 | Fable 5 | Anthropic | Chaotic month: Fable 5 and Mythos 5 launched June 9, briefly restricted days later, restored July 1. Anthropic made Sonnet 5 its default model June 30 |
| May 2026 | claude-opus-4-7 | Anthropic | Opus 4.8 launched May 28, near month’s end |
| April 2026 | claude-opus-4-7 | Anthropic (~$11-12M volume) | Opus 4.7 launched April 16, took the lead by mid-month |
| March 2026 | claude-opus-4-6-thinking (100%, 558,821) | Anthropic (99.3%, $16.22M) | Second consecutive Anthropic win. Anthropic opened March at 54% on Polymarket with Google at 24%, OpenAI at 12%, and xAI at 5.5% |
| Feb 2026 | claude-opus-4-6 (99%, $2.48M) | Anthropic (98%, $21.66M) | First non-Google winner. Claude launched Feb 5, took Arena #1 by Feb 6. Kalshi split: opus-4-6 at 85%, opus-4-6-thinking at 17% |
| Jan 2026 | gemini-3-pro ($1.54M) | Google ($28.97M) | Uncontested |
| Dec 2025 | gemini-3-pro ($471K) | Google ($36.33M) | Uncontested |
| Nov 2025 | gemini-3-pro ($810K) | Google (—) | Gemini 3 Pro launched mid-month, dethroned Gemini 2.5 Pro. gemini-2.5-pro led ~50% early |
| Oct 2025 | gemini-2.5-pro ($279K) | Google ($4.66M) | Volatile early on Kalshi, settled by mid-month |
| Sep 2025 | gemini-2.5-pro ($92K) | Google ($2.92M) | Uncontested on both |
| Aug 2025 | gemini-2.5-pro ($291K) | Google ($7.49M) | GPT-5 High hit 32% on Kalshi, OpenAI hit 73% on Polymarket mid-month before Google retook lead on both |
How to bet on AI prediction markets
Each company (or model, on Kalshi) gets its own contract. You pick Yes or No on whether that AI company will hold #1 at the end of the month. Only one can win, so in practice this is a winner-take-all contract. If you buy Anthropic Yes at 98¢ and Anthropic holds #1 on February 28, the contract pays $1.00 and every other company’s contract resolves No.
The contracts trade between 1¢ and 99¢. With only one day left in February and Anthropic at 97-99%, there isn’t much edge left to gain this month. The opportunity is in forward months where the outcome is less certain, particularly if GPT-5.3 or Grok 5 launches and disrupts the leaderboard.
The two platforms handle timing differently. Kalshi opens each month’s market on the 1st, so you can only trade the current month. Polymarket lists future months alongside the active one — March, June, and beyond are already tradeable. That means Polymarket lets you take a position on where the leaderboard will be months from now, while Kalshi is limited to the race in progress.
For a full comparison of where to trade, see our list of prediction market apps.
Anthropic sweeps June despite a mid-month export-control scare
Claude Opus 4.8, which launched May 28, carried Anthropic’s lead into June, and the June 9 launch of Claude Fable 5 and Claude Mythos 5 stretched it further. Both Kalshi and Polymarket resolved to Anthropic at the June 30 snapshot — Anthropic’s fifth consecutive monthly title.
The month wasn’t without drama. On June 12, U.S. export controls cut off access to Fable 5 and Mythos 5, and Anthropic suspended availability to comply. The controls were lifted June 30, and Anthropic restored access July 1 — the same day it made Claude Sonnet 5 its new default model. None of it moved the leaderboard; rivals never closed enough ground for the suspension to matter for market pricing.
An interesting twist: Anthropic didn’t just win “best AI model” for June, it also led the separate “second-best AI model” market, pricing in the low 90s% at one point. Traders were effectively betting that Claude’s model family held both the #1 and #2 spots on Arena simultaneously.
Polymarket volume on the June market ran around $16M as of mid-month — back in line with February’s $21.66M after March and April’s dip.
July has already delivered its own headline. Anthropic released Claude Opus 5 on July 24 — its fourth model in under two months, after Mythos 5, Fable 5, and Sonnet 5 — and it immediately topped Artificial Analysis’s newly rebased leaderboard. With the July market resolving July 31, Opus 5’s timing looks like it locked in another Anthropic month before rivals had a chance to respond.
What to watch this month
Several threads could shape next month’s market:
- Gemini 3.5 Pro is still MIA. Bloomberg reported July 16 that it’s running months behind Google’s internal schedule and falling short on coding benchmarks specifically. If it finally ships in August, it’s Google’s best (and possibly only) shot at pressuring Claude for the top spot.
- GPT-5.6 hasn’t gone wide yet. OpenAI previewed it June 26 to a restricted list of roughly 20 organizations. A broader public release in August would be the first real head-to-head test against Opus 5.
- Anthropic’s release cadence. Mythos 5, Fable 5, Sonnet 5, and Opus 5 all shipped within about seven weeks. Whether that pace continues into August matters a lot for how much room rivals get to close the gap.
- Regulatory risk isn’t fully behind us. The mid-June export-control suspension of Fable 5/Mythos 5 access shows government action can move fast on frontier models. Worth watching for any follow-on action, especially with “kill switch” legislation still circulating.
How do these markets resolve?
Kalshi and Polymarket both resolve against the Arena leaderboard, but they use different columns — which means they could theoretically resolve to different winners.
- Kalshi checks the “Rank (UB)” column on the text leaderboard with Style Control removed on the last day of the month. Tiebreaker goes to highest Arena Score, then most votes, then earliest release.
- Polymarket checks the “Arena Score” column on the Leaderboard tab at 12:00 PM ET on the last day of the month with style control off. Tiebreaker is alphabetical.
Rank (UB) reflects the upper bound of a model’s confidence interval, while Arena Score is the raw Elo. In a tight race where two models are separated by a few points, one platform could resolve to a different winner than the other — use our arbitrage calculator to compare pricing across platforms.
How Arena ranks AI models
Kalshi and Polymarket both resolve against the Arena leaderboard, formerly known as the LMSYS Chatbot Arena. Arena presents users with a prompt and two anonymous model responses side by side. The user picks which response is better without knowing which model produced it. Over 5.3 million of these blind comparisons have been collected, and each model receives an Elo rating derived from its win rate against other models in the pool.
Style Control is a leaderboard filter that adjusts for formatting bias — models that use more markdown, bullet points, or longer responses tend to score higher in raw comparisons even when the underlying reasoning is equivalent. With Style Control on, Arena isolates substance from presentation. Both Kalshi and Polymarket require Style Control off in their resolution criteria, meaning the raw Elo without formatting adjustment is what determines the winner.
Related AI prediction markets
- “Which companies will have a top-ranked AI model this year?” on Kalshi. An annual market with 12 active contracts. Google and Anthropic have already resolved Yes.
- “Will any AI model reach a 1550+ Overall Arena Score by December 31?” on Polymarket. Currently pricing around 18% — effectively a bet on whether the next generation of models (or Opus 5’s ceiling) clears that bar before year-end.
- “When will Gemini 3.5 Pro be released?” on Polymarket. Google’s next flagship has already slipped from a June target past a rumored July 17 date; recent reporting points to an August release, with Google reportedly having scrapped a near-complete build over quality issues. This is the market most likely to move if Google is going to make a real run at Anthropic.
- “When will GPT-5.6 reach general availability?” on Polymarket. OpenAI previewed the model to a restricted list of roughly 20 organizations back in June, and as of late July it’s still not broadly available — despite earlier market pricing implying a near-certain public release by July 31.
- EU AI Act compliance deadline arrives August 2, 2026 — just days away. If a lab delays a release to meet compliance requirements, that’s one fewer shot at the leaderboard in a given month.
Liquidity varies significantly across these markets. Polymarket’s monthly AI market routinely draws $20-36 million per month, making it one of the deeper non-political betting markets on the platform. Kalshi’s monthly markets are thinner at $2-3 million but growing — February volume nearly doubled January. The annual and milestone markets on either site are much smaller, typically under $1 million, so larger positions may face slippage.
The fee structures are quite different between Kalshi and Polymarket and can affect net returns, particularly on lower-volume contracts. If you’re new to betting on Kalshi or Polymarket, you will also want to learn how order books work and market resolution before making your first trade.
FAQ
What does “best AI model” mean in these markets?
The company (or model) that holds the top rank on the Arena text leaderboard at the end of the month. Kalshi uses the Rank (UB) column, Polymarket uses Arena Score.
Can Kalshi and Polymarket resolve to different winners?
In theory, yes. They use different columns from the same leaderboard. In practice, the #1 model under Rank (UB) and Arena Score has been the same, but a tight race could produce a split.
Why is Polymarket’s volume so much higher than Kalshi’s?
Polymarket generally carries more liquidity on AI-related markets and draws a larger crypto-native trading audience. Kalshi is a regulated US exchange that has been expanding internationally, while Polymarket US continues to roll out to domestic traders.
Why is Anthropic such a heavy favorite right now?
The gap has only widened since spring. Anthropic has won every monthly market since February, and its release cadence has outpaced rivals: Claude Opus 4.8 (May 28), Fable 5 and Mythos 5 (June 9), Sonnet 5 (June 30), and Opus 5 (July 24) — four models in under two months. Opus 5 immediately topped Artificial Analysis’s leaderboard on release. Google’s Gemini 3.5 Pro, meanwhile, has been delayed repeatedly and remains unreleased, and OpenAI’s GPT-5.6 is still limited to a small preview group rather than broadly available. Until one of those changes, there’s little on the horizon to close the gap.
How often do these markets run?
Monthly on Kalshi and Polymarket. New contracts open as the prior month’s market approaches resolution. Polymarket has results going back to at least August 2025, Kalshi to at least August 2025 as well.