List your company

Legal Moves Prediction on Chess Reasoning Dataset Legal Moves

96IoUgpt-oss-120b (Medium)
13.8435.1756.577.83Apr 6, 2026

Evaluation results

MethodLinks
gpt-oss-120b (Medium)Training Stage=Baseline2026.0496
SFT8 + SFT9 XLTraining Stage=RL2026.0487
SFT8 XLTraining Stage=RL2026.0479
SFT9 XLTraining Stage=RL2026.0475
SFT8 [BM - All]Training Stage=RL2026.0467
SFT6 [BM]Training Stage=RL2026.0466
SFT7 [BL]Training Stage=RL2026.0463
SFT9 [BL - All]Training Stage=RL2026.0459
SFT3 [VABP]Training Stage=RL2026.0458
SFT4 [FBA]Training Stage=RL2026.0458
SFT5 [GS]Training Stage=RL2026.0457
SFT8 + SFT9 XLTraining Stage=SFT2026.0452
SFT4 [FBA]Training Stage=SFT2026.0449
SFT9 XLTraining Stage=SFT2026.0447
SFT7 [BL]Training Stage=SFT2026.0446
SFT8 XLTraining Stage=SFT2026.0446
SFT2 [RSA]Training Stage=RL2026.0444
SFT8 [BM - All]Training Stage=SFT2026.0444
Llama 4 MaverickTraining Stage=Baseline2026.0443
SFT6 [BM]Training Stage=SFT2026.0442
SFT3 [VABP]Training Stage=SFT2026.0441
SFT9 [BL - All]Training Stage=SFT2026.0440
SFT2 [RSA]Training Stage=SFT2026.0437
SFT5 [GS]Training Stage=SFT2026.0437
Qwen2.5 7B-InstructTraining Stage=Baseline2026.0426
SFT1 [RSPM]Training Stage=SFT2026.0426
SFT1 [RSPM]Training Stage=RL2026.0417