AIME2025

Fresh

Problems from the American Invitational Mathematics Examination (AIME) 2025-I & II.

Type: RL Env
Publisher: General Reasoning
Tags: Mathematical Reasoning
Runtime: ORS
License: unknown
Size: 30 tasks
Published: Jan 2026
Canonical: openreward.ai/GeneralReasoning/AIME2025

Cite

Notes

Only stored in your browser.

Attribution

README: openreward.ai/GeneralReasoning/AIME2025
Scores: OpenReward

Attribution policy →

Public scores on this env

14

15 vf-eval reports across 14 models

1Grok 4 Heavy (with python)xAI100 2Claude Sonnet 4.5Anthropic100 3GPT-5.2OpenAI100 4Step 3.5 Flash (parallel thinking)99.9 5Claude Opus 4.6Anthropic99.79 6Grok 4xAI98.8 7Step 3.5 Flash97.3 8Kimi K2.5Moonshot AI96.1 9Gemini 3 ProGoogle (Alphabet Inc.)95 10o4 MiniOpenAI92.7 11o3OpenAI88.9 12o3 MiniOpenAI86.5 13MiniMax M2.5Minimax86.3 14Qwen 3 Coder NextAlibaba83.07

Open the scoring view →