programbench/20260717_mini-v2.4.5_opus-4.8-xhigh
Viewer • Updated • 1 • 11
ProgramBench
university
AI & ML interests
None defined yet.
Recent Activity
View all activity
Organization Card
ProgramBench: Can Language Models Rebuild Programs From Scratch?
Given only a compiled binary and its documentation, AI agents must architect and implement a complete codebase that reproduces the original program's behavior. ProgramBench evaluates this capability across 200 real-world open-source projects spanning Rust, Go, C, C++, Haskell, and Java.
Links
models 0
None public yet
datasets 15
programbench/20260717_mini-v2.4.5_glm-5.2
Viewer • Updated • 1 • 12
programbench/20260508_mini-v2.2.6_gpt-5-5-xhigh
Viewer • Updated • 1 • 522
programbench/20260507_mini-v2.2.6_gpt-5-5
Viewer • Updated • 1 • 429
programbench/20260507_mini-v2.2.6_gpt-5-5-high
Viewer • Updated • 1 • 479
programbench/20260430_mini-v2.2.6_opus-4-6
Viewer • Updated • 1 • 395
programbench/20260429_mini-v2.2.6_sonnet-4-6
Viewer • Updated • 1 • 441
programbench/20260429_mini-v2.2.6_opus-4-7
Viewer • Updated • 1 • 455
programbench/20260429_mini-v2.2.6_haiku-4-5
Viewer • Updated • 1 • 447
programbench/20260429_mini-v2.2.6_gpt-5-mini
Viewer • Updated • 1 • 488