# Competitive Mathematical Reasoning on OlympiadBench (pass@50) Source: [Competitive Mathematical Reasoning on OlympiadBench (pass@50)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:57:36.392Z Evaluation of model performance on competitive mathematical reasoning problems using the OlympiadBench dataset, measured by pass@50. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Competitive Mathematical Reasoning | | Dataset (including any stated split) | OlympiadBench | | Metric | Pass@50 | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 2 of 2 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Pass@50](): Higher is better; unit: PERCENT. ## Recorded results | Method | Pass@50 | Result date | Report paper | | --- | --- | --- | --- | | Gemma-3-27b-it | 43.52 | Not recorded | 68fa7bdd1e209131cea8ae2c | | Gemma-3-27b-it | 33.19 | Not recorded | 68fa7bdd1e209131cea8ae2c | ### Gemma-3-27b-it Pass@50: 43.52. Result ID: 699a69f65698665ec4f2532e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sampling Strategy":"GUIDEDSAMPLING"}. This record does not assert independent reproduction. ### Gemma-3-27b-it Pass@50: 33.19. Result ID: 699a69f65698665ec4f2532d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sampling Strategy":"Repeated Sampling"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.