Skip to main content
ModelScale

external benchmark

ProgramBench leaderboard

Vals ProgramBench. Every model the catalog carries a published ProgramBench value for, ranked by that value.

Categoryexternal
MeasureAccuracy score
TasksProgram reconstruction tasks
DifficultyCleanroom software engineering
Published byProgramBench

Can language models rebuild programs from scratch?

No model in the catalog has a published ProgramBench score.

The benchmark is defined by ProgramBench, but the catalog carries no value for it yet. An absent value is shown as absent here rather than as a zero.

All leaderboards