Share your thoughts, 1 month free Claude Pro on usSee more

Mathematical Reasoning on OlympiadBench (accuracy)

0.7659Accuracy

JustRL-Nemotron

Updated 25d ago

Evaluation Results

Method
JustRL-Nemotron 2025.12	0.7659	-
QuestA 2025.12	0.7228	-
Backbone 2025.12	0.717	-
PMPO-7B (Ours) [R1-Distill] 2026.01	0.642	-
GMPO-7B [R1-Distill] 2026.01	0.625	-
Budget Allocation w/ V0 2026.02	0.5634	-
Dr.GRPO-7B [R1-Distill] 2026.01	0.557	-
Budget Allocation w/o V0 2026.02	0.5497	-
GRPO 2026.02	0.5453	-
R1-Distill-Qwen-7B-R-TAP 2026.03	0.538	-
OpenReasoner-Zero-7B@8k 2026.01	0.479	-
OpenReasoner-Zero-7B @ 8k 2026.03	0.479	-
R1-Distill-Qwen-7B @ 8k 2026.03	0.477	-
Oat-Zero-7B-R-TAP 2026.03	0.468	-
WIST 2026.03	0.468	-
PMPO-7B (Ours) 2026.01	0.467	-
SPICE 2026.03	0.467	-
SPICE 2026.03	0.453	-
R-Zero 2026.03	0.452	-
R-Zero 2026.03	0.452	-
WIST 2026.03	0.441	-
OpenReasoner-Zero-7B@3k 2026.01	0.44	-
OpenReasoner-Zero-7B @ 3k 2026.03	0.44	-
GMPO-7B 2026.01	0.436	-
Base Model 2026.03	0.431	-
Oat-Zero-1.5B-R-TAP 2026.03	0.428	-
Qwen2.5-Math-7B-Instruct 2026.03	0.427	-
GPG-7B 2026.01	0.424	-
R1-Distill-Qwen-1.5B-R-TAP @ 8k 2026.03	0.424	-
Dr.GRPO-7B 2026.01	0.41	-
Oat-Zero-7B 2026.03	0.41	-
SPICE 2026.03	0.41	-
PRIME-Zero-7B 2026.01	0.409	-
Eurus-7B 2026.01	0.409	-
PRIME-Zero-7B 2026.03	0.409	-
Base Model 2026.03	0.405	-
SimpleRL-Zero-7B 2026.01	0.403	-
SimpleRL-Zero-7B 2026.03	0.403	-
Qwen2.5-Math-1.5B-Instruct 2026.01	0.402	-
Qwen2.5-Math-1.5B-Instruct 2026.03	0.402	-
WIST 2026.03	0.4	-
R-Zero 2026.03	0.395	-
PMPO-1.5B (Ours) 2026.01	0.393	-
GMPO-1.5B 2026.01	0.387	-
Dr.GRPO-1.5B 2026.01	0.376	-
Oat-Zero-1.5B 2026.03	0.376	-
R1-Distill-Qwen-1.5B @ 8k 2026.03	0.358	-
Qwen2.5-Math-7B 2026.03	0.347	-
Base Model 2026.03	0.302	-
Qwen2.5-Math-1.5B 2026.01	0.284	-
Qwen2.5-Math-1.5B 2026.03	0.284	-
GRPO 2025.11	0.26	-
OBLR-PO 2025.11	0.26	-
R-Zero 2026.03	0.259	-
WIST 2026.03	0.255	-
ReMax 2025.11	0.253	-
SPICE 2026.03	0.249	-
RLOO 2025.11	0.247	-
OBLR-PO 2025.11	0.241	-
PPO 2025.11	0.238	-
GRPO 2025.11	0.235	-
R1-Distill-Qwen-7B @ 3k 2026.03	0.231	-
RLOO 2025.11	0.226	-
ReMax 2025.11	0.224	-
PPO 2025.11	0.22	-
R1-Distill-Qwen-1.5B @ 3k 2026.03	0.173	-
SPICE 2026.03	0.173	-
R-Zero 2026.03	0.172	-
Base Model 2026.03	0.17	-
Qwen2.5-Math-7B 2026.01	0.166	-
WIST 2026.03	0.151	-
Base Model 2026.03	0.11	-
Base 2025.09	-	0.5096
GRPO 2025.09	-	0.4444
λ-GRPO 2025.09	-	0.5348
GRPO 2025.09	-	0.52
λ-GRPO 2025.09	-	0.5378
Base 2025.09	-	0.0563
GRPO 2025.09	-	0.0607
λ-GRPO 2025.09	-	0.0622
GRPO 2025.09	-	0.0533
λ-GRPO 2025.09	-	0.0489