Atlas

Benchmarks

← All benchmarks

ChessBench — Hanging Piece

Games · 2026-03-22

ChessBench Hanging Piece contains 30 fixed non-mating Lichess tactics labeled with the hanging-piece theme. Models are scored only when their returned UCI move line exactly matches the expected line; the track is evenly stratified across three puzzle-rating buckets.

Top models (higher is better)

ModelScore
GPT-6.1 Sol63.3
Claude Opus 5.560.0
Gemini 3.8 Flash60.0
Gemini 4 Argon60.0
Claude Fable 5.143.3
GPT-6 Sol43.3
Claude Sonnet 5.540.0
Gemini 3 Flash Preview30.0
Claude Haiku 5.526.7
Grok 4.626.7
GPT-6 Luna16.7
Haiku 4.50.0
Loading Atlas data…