Highest Google Gemini score on Humanity’s Last Exam in 2026?

According to the latest prediction market data for the query “Highest Google Gemini score on Humanity’s Last Exam in 2026?”, traders have formed a strong consensus.

Currently, 60%+ is dominating the market with an overwhelming 19% chance of winning. 65%+ follows in second place at 6%, while 70%+ sits in third with 4%. The betting volume for this specific market has already reached $137.9K, reflecting intense industry interest.

To better assess where each potential outcome stands, the market can be segmented into three distinct trading tiers based on implied probability and contract pricing:

🥇 Tier 1: The Dominant Leader

• 60%+ (19%): Currently commanding the highest probability, 60%+ is heavily favored by the order book. Traders looking to back this outcome face a “Buy Yes” contract price of 19¢, signaling a high degree of market conviction. This contract alone has generated $22.4K in volume.

🥈 Tier 2: The Primary Challengers

• 65%+ (6%): Positioned as the most viable alternative, 65%+ maintains a 6% chance of resolving true. Its “Buy Yes” shares currently trade at 6¢.
• 70%+ (4%): Sitting in third place with a 4% probability, the market shows measured skepticism toward 70%+, treating it as an outside wildcard unless momentum shifts.

This market will resolve to "Yes" if any Google Gemini model achieves at least the specified accuracy on Humanity’s Last Exam by December 31, 2026, 11:59 PM ET. Otherwise, this market will resolve to "No".

For resolution, “accuracy” refers solely to the value labeled “HLE Accuracy” or a clear equivalent metric if the data’s presentation or terminology is restructured, regardless of the model’s Calibration Error or any other displayed metric.

The resolution source will be the official Humanity’s Last Exam leaderboard at https://agi.safe.ai/. If the resolution source becomes unavailable during the listed timeframe, this market will remain open to allow the relevant data to become available again. If the source remains unavailable after the end of the listed timeframe or is otherwise confirmed to be permanently unavailable, official Humanity’s Last Exam results published elsewhere may be used. If no official alternative source is available, this market will resolve to "No".

While human consensus and speculative volume shape the broader prediction market, our quantitative algorithms offer a data-driven counter-perspective. By analyzing fundamental signals, underlying trends and historical distributions, our AI Valuation model calculates an independent “Fair Value” probability for each outcome.

Comparing this Fair Value against the current Trade Value uncovers major disparities — known as the Expected Value (EV) Gap. Contracts with a positive EV Gap represent statistically underpriced outcomes, whereas a negative EV Gap flags a potential market overreaction.

Top AI Alpha & Mispriced Arbitrage Opportunities

Based on the latest data model run, several key contracts stand out with significant deviations:

• The Best Value Play (Highest EV) Our model identifies 70%+ as the premium value opportunity on the board. While the market only assigns it a 4% trading probability, our AI’s Fair Value assessment sits at 14.7% — yielding an impressive +10.7% EV Gap.
• Under-the-Radar Dark Horses Other notable discrepancies include 65%+ (EV Gap: +3.9%). These long-tail opportunities are heavily discounted by the live order books despite stronger statistical backing from our predictive model.