Checkpoints and eval traces from "Training frontier knowledge work agents: A 397B open recipe with SkyRL"
AI & ML interests
None defined yet.
Recent Activity
datasets 9
mercor/OpenWorldEngineer
Viewer • Updated • 26 • 468 • 2
mercor/apex-accounting-harbor
Viewer • Updated • 10 • 2.65k • 3
mercor/apex-agents-v1.1
Viewer • Updated • 240 • 11.3k • 13
mercor/ApexAgentsRecipe-TBench2_1-EvalTraces
Updated • 705
mercor/ApexAgentsRecipe-ApexAgents480-EvalTraces
Updated • 527 • 1
mercor/apex-agents
Benchmark • Updated • 480 • 207k • 223
mercor/APEX-v1-extended
Benchmark • Updated • 100 • 3.39k • 17
mercor/ACE
Benchmark • Updated • 592 • 421 • 7
mercor/CL-bench-samples
Viewer • Updated • 4 • 115 • 1