Timeline of MBZUAI Institute of Foundation Models

k2-think-v2 scores 14.98% on Terminal-Bench

k2-think-v2 scores 25.44% on τ²-bench

k2-think-v2 scores 62.79% on IFBench

k2-think-v2 scores 10.1% on Humanity's Last Exam

k2-think-v2 scores 71.31% on GPQA

k2-think-v2 scores 11.5 on Artificial Analysis Intelligence Index

k2-v2 scores 9.85% on Terminal-Bench

k2-v2 scores 27.78% on τ²-bench

k2-v2 scores 60.14% on IFBench

k2-v2 scores 10.52% on Humanity's Last Exam

k2-v2 scores 68.08% on GPQA

k2-v2 scores 9.87 on Artificial Analysis Intelligence Index

k2-horizon-375b-a23b scores 71.91% on Terminal-Bench

k2-horizon-375b-a23b scores 42.94% on SciCode

k2-horizon-375b-a23b scores 31.97% on Humanity's Last Exam

k2-horizon-375b-a23b scores 87.27% on GPQA

k2-horizon-375b-a23b scores 30.7 on Artificial Analysis Intelligence Index

k2-v2-low scores 4.55% on Terminal-Bench

k2-v2-low scores 20.76% on τ²-bench

k2-v2-low scores 40.95% on IFBench

k2-v2-low scores 3.61% on Humanity's Last Exam

k2-v2-low scores 54.14% on GPQA

k2-v2-low scores 7.31 on Artificial Analysis Intelligence Index

k2-v2-medium scores 8.33% on Terminal-Bench

k2-v2-medium scores 24.85% on τ²-bench

k2-v2-medium scores 55.1% on IFBench

k2-v2-medium scores 4.36% on Humanity's Last Exam

k2-v2-medium scores 59.8% on GPQA

k2-v2-medium scores 9.02 on Artificial Analysis Intelligence Index