| mathematical-reasoning-on-hmmt-feb-2026 | K2-Horizon-7B | #8 | Accuracy: 73.3 |
| agents-on-toolathlon-verified-108-tasks | K2-Horizon-375B-A23B | #10 | Pass@1: 0.653 |
| agents-on-automationbench-public | K2-Horizon-375B-A23B | #12 | Success Rate: 25.3 |
| code-generation-on-scicode | K2-Horizon-7B | #23 | Accuracy: 31.6 |
| deep-research-agents-on-browsecomp | K2-Horizon-375B-A23B | #26 | Accuracy: 72.8 |
| question-answering-on-humanity-s-last-exam-hle | K2-Horizon-7B | #31 | Accuracy: 18.6 |
| code-generation-on-swe-bench-verified | K2-Horizon-7B | #39 | Accuracy: 70.6 |
| code-generation-on-terminal-bench-2-1 | K2-Horizon-7B | #44 | Accuracy: 39.1 |