Today for AI
HOT RADAR
llmHEAT 13.5°

AI CURATED HOT STORY · 11/11/2026

Sixteen LLMs Play TIS-100 and Set 133 New Frontier Points

1 reports archived1 independent sourcesupdated 11/11/2026, 08:37:05
Synthesis & Latest Updates
1 Sources Cross-Validated

Wincon Research had 16 frontier and open-weight LLMs attempt all 47 TIS-100 assembly puzzles with the community leaderboard withheld; the models produced 133 new frontier points, new records on 18 puzzles, and 3 outright cycle records. Human players then picked up and improved most of the generated programs within days, showing the outputs were genuinely usable engineering artifacts.

LATEST/Wincon Research had 16 frontier and open-weight LLMs attempt all 47 TIS-100 assembly puzzles with the community leaderboard withheld; the models produced 133 new frontier points, new records on 18 puzzles, and 3 outright cycle records. Human players then picked up and improved most of the generated programs within days, showing the outputs were genuinely usable engineering artifacts.

Hourly heat curve

4 fully observed hours
Now
13.5
Peak
14.209:00
24h change
–

The line compares only sources observed throughout; gap hours are interpolated to keep the trend continuous.

Coverage timeline

Total 1 reports · Latest first
  1. Hacker News AIT2·72 pts

    Sixteen LLMs Play TIS-100 and Set 133 New Frontier Points

    • All 16 models attempted the 47 puzzles, yielding 133 new frontier points, records on 18 puzzles, and 3 outright cycle records
    • TIS-100 scores on cycles, nodes, and instructions, creating a Pareto frontier that tests multi-objective tradeoff reasoning
    • Human players improved most model programs within days, indicating readable and editable output that still trails top human optimization