Math Reasoning on AIME 2024 (Mean Wall-Clock Time)
440.5Mean Wall-Clock Time (s)lag-kv
Evaluation Results
| Method | Links | |
|---|---|---|
| lag-kvBudget=81922026.06 | 440.5 | |
| band-adaptiveBudget=81922026.06 | 493.3 | |
| kv-valBudget=81922026.06 | 497.7 | |
| lag-kvBudget=40962026.06 | 499.7 | |
| lag-kvBudget=10242026.06 | 505 | |
| band-adaptiveBudget=40962026.06 | 521.6 | |
| kv-valBudget=40962026.06 | 523.9 | |
| lag-kvBudget=20482026.06 | 533 | |
| EPIKVBudget=81922026.06 | 538.5 | |
| band-adaptiveBudget=5122026.06 | 547.4 | |
| kv-valBudget=5122026.06 | 547.9 | |
| hs-varianceBudget=81922026.06 | 551.1 | |
| band-adaptiveBudget=10242026.06 | 551.1 | |
| kv-valBudget=10242026.06 | 553.1 | |
| band-adaptiveBudget=20482026.06 | 556.1 | |
| kv-valBudget=20482026.06 | 560.4 | |
| hs-varianceBudget=40962026.06 | 561.7 | |
| EPIKVBudget=40962026.06 | 565 | |
| EPIKVBudget=10242026.06 | 601.3 | |
| hs-varianceBudget=10242026.06 | 604.3 | |
| EPIKVBudget=20482026.06 | 614.8 | |
| hs-varianceBudget=20482026.06 | 615.3 | |
| hs-varianceBudget=5122026.06 | 617.5 | |
| EPIKVBudget=5122026.06 | 630.1 | |
| thinKVBudget=81922026.06 | 721 | |
| kv-keyBudget=81922026.06 | 732.1 | |
| thinKVBudget=40962026.06 | 746.5 | |
| noneBudget=20482026.06 | 762 | |
| hybrid-seg-hsBudget=81922026.06 | 763 | |
| thinKVBudget=20482026.06 | 770.5 | |
| kv-keyBudget=40962026.06 | 782.9 | |
| hybrid-seg-hsBudget=40962026.06 | 783.2 | |
| thinKVBudget=10242026.06 | 788.1 | |
| thinKVBudget=5122026.06 | 849 | |
| h2oBudget=81922026.06 | 864.5 | |
| h2oBudget=40962026.06 | 865.7 | |
| lag-kv-keyBudget=5122026.06 | 874.4 | |
| lag-kv-keyBudget=81922026.06 | 879.7 | |
| kv-keyBudget=20482026.06 | 884.6 | |
| attn×hsBudget=40962026.06 | 884.9 | |
| raasBudget=5122026.06 | 885.2 | |
| attn×hsBudget=81922026.06 | 889.8 | |
| kv-keyBudget=10242026.06 | 891.5 | |
| kv-keyBudget=5122026.06 | 893.8 | |
| raasBudget=10242026.06 | 899.2 | |
| lag-kv-keyBudget=40962026.06 | 913.8 | |
| raasBudget=20482026.06 | 921.6 | |
| hybrid-seg-hsBudget=10242026.06 | 927.8 | |
| hybrid-seg-hsBudget=20482026.06 | 933.9 | |
| hybrid-seg-hsBudget=5122026.06 | 944.2 | |
| h2oBudget=5122026.06 | 963.2 | |
| lag-kv-keyBudget=10242026.06 | 974.9 | |
| h2oBudget=10242026.06 | 981.1 | |
| lag-kv-keyBudget=20482026.06 | 998.3 | |
| h2oBudget=20482026.06 | 1,007 | |
| lag-kvBudget=5122026.06 | 1,013.6 | |
| attn×hsBudget=20482026.06 | 1,021.2 | |
| raasBudget=81922026.06 | 1,238.7 | |
| raasBudget=40962026.06 | 1,285.2 | |
| attn×hsBudget=5122026.06 | 1,327.5 | |
| attn×hsBudget=10242026.06 | 1,352.5 |