GPT5.4-N TrueSkill μ=22.6±2.7

GPT-5.4 Nano TrueSkill rating

Source paper

extracted_from

(2026) · Robert Müller · Clemens Müller

cosine ≥ 0.65 · no typed edge

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

G2.5-FL TrueSkill μ=20.5±2.9finding0.863
Gemini 2.5 Flash Lite, lowest TrueSkill rating
DS-v3.2 TrueSkill μ=23.9±2.7finding0.846
DeepSeek v3.2 TrueSkill rating
SetRaceAgent TrueSkill μ=27.3±3.3finding0.810
fourth-highest TrueSkill rating
G3.1-FL TrueSkill μ=28.0 ± 2.9, 44.9% win ratefinding0.807
Second-best LLM, competitive with TrackerAgent.
EconomyAgent TrueSkill μ=22.1±3.5finding0.793
EconomyAgent TrueSkill rating
Haiku TrueSkill μ=21.8±2.8finding0.791
Claude Haiku 4.5 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)finding0.790
Top LLM performance with high win rate and large score.
Sonnet 4.5 TrueSkill μ=26.4 ± 4.9 (n=14, 35.7% win rate)finding0.784
Mid-field performance with larger uncertainty due to small sample.