ResearchBenchmarksMath Reasoning on HMMT November 2025 edition (test)Follow4,025Optimization Token UsageETGPO3,720.845,773.927,8279,880.08Feb 1, 2026Evaluation ResultsMethodMethodLinksOptimization Token UsageETGPOBackbone model=GPT-4.1...Backbone model=GPT-4.1-mini, Optimizer=GPT-4.12026.024,025GEPABackbone model=GPT-4.1...Backbone model=GPT-4.1-mini, Optimizer=GPT-4.12026.0210,249MIPROv2Backbone model=GPT-4.1...Backbone model=GPT-4.1-mini, Optimizer=GPT-4.12026.0211,629