BenchmarkAIME 2025
AIME25
AIME25 is the 2025 problem set of the American Invitational Mathematics Examination, commonly used to evaluate the mathematical reasoning ability of large models and serving as one of the LLM benchmarks.
Timeline (last 90 days)
Oct 30
Unsloth ran reproduction tests on ISTA's GSQ-RCO GGUFs and failed to obtain the results ISTA reported; on AIME25, ISTA claimed 100% accuracy recovery while Unsloth measured only about 91.88%.
Unverified50%