TrackerAgent TrueSkill μ=28.7±3.6 in 98 canonical games

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• TrackerAgent TrueSkill μ=28.7 ± 3.6, 53.6% win ratefinding0.887
  Best code agent outperforming six of seven LLMs.
• TrackerAgent μ=30.8 in 172-game mixed slicefinding0.841
  TrackerAgent uses card counting to achieve high rating, a capability no LLM replicates
• SetRaceAgent TrueSkill μ=27.3±3.3finding0.833
  fourth-highest TrueSkill rating
• EconomyAgent TrueSkill μ=22.1±3.5finding0.817
  EconomyAgent TrueSkill rating
• G3-F TrueSkill μ=30.1 ± 3.3, 72.9% wins, median score 5,250 on combined-comp1 slice (n=98 canonical games)finding0.816
  Top LLM performance with high win rate and large score.
• TrackerAgent win rate 53.6% in 98 canonical gamesfinding0.812
  TrackerAgent won over half of the combined-comp1 games.
• DS-v3.2 TrueSkill μ=23.9±2.7finding0.810
  DeepSeek v3.2 TrueSkill rating
• G3.1-FL TrueSkill μ=28.0 ± 2.9, 44.9% win ratefinding0.795
  Second-best LLM, competitive with TrackerAgent.