List your company

Diagnostic on Held-out games (test)

6.07Quality ScoreBG-Critic
1.58762.75133.9155.0787Jun 1, 2026

Evaluation results

MethodLinks
BG-CriticStage 2 multi-task training=Included2026.066.07
GPT-5.42026.063.92
Gemini-3.1-Flash2026.062.64
Qwen3.5-397B2026.062.39
Qwen3.5-27BBackbone Status=untuned2026.061.99
BG-CriticStage 2 multi-task training=Excluded2026.061.76