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
Grok 4.1 Fast53.3
Grok 4.553.3
Claude Fable 550.0
Gemini 3.1 Pro Preview50.0
Grok 4.2043.3
Grok Build 0.143.3
Gemini 3.5 Flash40.0
Grok 4.336.7
Gemini 3.1 Flash Image Preview33.3
Gemini 3 Flash Preview30.0
GPT-5.6 Terra30.0
GPT-5.423.3
Qwen3.6 Plus Preview23.3
GPT-5.520.0
GPT-5.6 Sol20.0
GLM-5.116.7
Opus 4.713.3
Opus 4.813.3
GPT-5.6 Luna Pro10.0
Qwen3.6 Plus (2026-04-02)10.0
Opus 4.66.7
Sonnet 4.66.7
GLM-56.7
GLM-5.23.3
Haiku 4.50.0
Sonnet 50.0
Gemini 2.5 Pro0.0
Loading Atlas data…