List your company

Mathematical Reasoning on AIME (Token Savings, Relative Accuracy Shift)

93.7Token SavingsGuardian
-3.5421.70546.9572.195Jan 6, 2026

Evaluation results

MethodLinks
GuardianModel Backbone=DSR, Sparsity Algorithm=H2O, Cache Budget=1282026.0193.70
GuardianModel Backbone=DSR, Sparsity Algorithm=Sink, Cache Budget=1282026.0192.50
GuardianModel Backbone=DSR, Sparsity Algorithm=H2O, Cache Budget=2562026.0191.20
GuardianModel Backbone=DSL, Sparsity Algorithm=H2O, Cache Budget=1282026.01910
GuardianModel Backbone=DSR, Sparsity Algorithm=Sink, Cache Budget=2562026.0190.9-0.5
GuardianModel Backbone=DSL, Sparsity Algorithm=H2O, Cache Budget=2562026.0187.90
GuardianModel Backbone=DSR, Sparsity Algorithm=Sink, Cache Budget=5122026.0187.8-0.5
GuardianModel Backbone=DSL, Sparsity Algorithm=H2O, Cache Budget=5122026.0187.40
GuardianModel Backbone=DSR, Sparsity Algorithm=H2O, Cache Budget=5122026.0186.40
GuardianModel Backbone=DSL, Sparsity Algorithm=H2O, Cache Budget=10242026.0183.60
GuardianModel Backbone=DSR, Sparsity Algorithm=Sink, Cache Budget=10242026.0179.90
GuardianModel Backbone=DSR, Sparsity Algorithm=H2O, Cache Budget=10242026.0179.60
GuardianModel Backbone=DSL, Sparsity Algorithm=Sink, Cache Budget=2562026.0172.60
GuardianModel Backbone=DSL, Sparsity Algorithm=Sink, Cache Budget=1282026.01710
GuardianModel Backbone=DSL, Sparsity Algorithm=Sink, Cache Budget=5122026.0167.60
GuardianModel Backbone=DSL, Sparsity Algorithm=Sink, Cache Budget=10242026.01650
GuardianModel Backbone=DSL, Sparsity Algorithm=Quest, Cache Budget=5122026.0153.90
GuardianModel Backbone=DSL, Sparsity Algorithm=Quest, Cache Budget=10242026.0148.2-1
GuardianModel Backbone=DSL, Sparsity Algorithm=Quest, Cache Budget=2562026.0144.90
GuardianModel Backbone=Qwe, Sparsity Algorithm=H2O, Cache Budget=2562026.0142.70
GuardianModel Backbone=Qwe, Sparsity Algorithm=H2O, Cache Budget=1282026.0139.50
GuardianModel Backbone=Qwe, Sparsity Algorithm=Sink, Cache Budget=1282026.0136.80.5
GuardianModel Backbone=DSR, Sparsity Algorithm=infLLM, Cache Budget=10242026.0133.20.5
GuardianModel Backbone=Qwe, Sparsity Algorithm=Sink, Cache Budget=2562026.0131.70
GuardianModel Backbone=Qwe, Sparsity Algorithm=H2O, Cache Budget=5122026.0131.60
GuardianModel Backbone=DSR, Sparsity Algorithm=infLLM, Cache Budget=5122026.0124.10
GuardianModel Backbone=DSL, Sparsity Algorithm=Quest, Cache Budget=1282026.0123.20.5
GuardianModel Backbone=DSR, Sparsity Algorithm=Quest, Cache Budget=5122026.0120.30
GuardianModel Backbone=DSR, Sparsity Algorithm=Quest, Cache Budget=2562026.0119.7-0.5
GuardianModel Backbone=DSR, Sparsity Algorithm=Quest, Cache Budget=1282026.0118.4-0.5
GuardianModel Backbone=DSR, Sparsity Algorithm=infLLM, Cache Budget=2562026.0116.80.5
GuardianModel Backbone=Qwe, Sparsity Algorithm=Sink, Cache Budget=5122026.0116.10
GuardianModel Backbone=DSL, Sparsity Algorithm=infLLM, Cache Budget=10242026.0116-2
GuardianModel Backbone=DSR, Sparsity Algorithm=Quest, Cache Budget=10242026.0114.5-1
GuardianModel Backbone=DSR, Sparsity Algorithm=infLLM, Cache Budget=1282026.0114.10
GuardianModel Backbone=Qwe, Sparsity Algorithm=H2O, Cache Budget=10242026.01110
GuardianModel Backbone=DSL, Sparsity Algorithm=infLLM, Cache Budget=5122026.0110-0.5
GuardianModel Backbone=Qwe, Sparsity Algorithm=Sink, Cache Budget=10242026.0160
GuardianModel Backbone=DSL, Sparsity Algorithm=infLLM, Cache Budget=1282026.014.80
GuardianModel Backbone=DSL, Sparsity Algorithm=infLLM, Cache Budget=2562026.013.90
GuardianModel Backbone=Qwe, Sparsity Algorithm=infLLM, Cache Budget=10242026.012.70
GuardianModel Backbone=Qwe, Sparsity Algorithm=Quest, Cache Budget=10242026.011.60.5
GuardianModel Backbone=Qwe, Sparsity Algorithm=Quest, Cache Budget=1282026.011.40
GuardianModel Backbone=Qwe, Sparsity Algorithm=Quest, Cache Budget=2562026.011.10
GuardianModel Backbone=Qwe, Sparsity Algorithm=infLLM, Cache Budget=5122026.011.10
GuardianModel Backbone=Qwe, Sparsity Algorithm=infLLM, Cache Budget=2562026.010.80
GuardianModel Backbone=Qwe, Sparsity Algorithm=Quest, Cache Budget=5122026.010.60
GuardianModel Backbone=Qwe, Sparsity Algorithm=infLLM, Cache Budget=1282026.010.20