https://programbench.com/task/mgechev__revive.201451e/
mgechev/revive — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
reviveprogrambench
https://programbench.com/task/ogham__dog.721440b/
ogham/dog — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
oghamdogprogrambench
https://programbench.com/task/samtools__samtools.aa823b5/
samtools/samtools — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
samtoolsprogrambench
https://programbench.com/task/abishekvashok__cmatrix.5c082c6/
abishekvashok/cmatrix — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
cmatrixprogrambench
https://programbench.com/task/ammarabouzor__tui-journal.2b4540d/
AmmarAbouZor/tui-journal — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
tuijournalprogrambench
https://programbench.com/task/stathissideris__ditaa.f2286c4/
stathissideris/ditaa — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
ditaaprogrambench
https://www.simplenews.ai/news/facebook-researchs-programbench-shows-0percent-of-ai-models-can-rebuild-complete-programs-from-scratch-o1a0
Facebook Research's ProgramBench Shows 0% of AI Models Can Rebuild Complete Programs From Scratch |...
May 7, 2026 - Facebook Research's new ProgramBench benchmark reveals that 0% of evaluated AI models can fully reconstruct complete working programs from binaries and...
facebook researchai modelscomplete programsfrom scratch
https://programbench.com/task/google__brotli.b3dc9cc/
google/brotli — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
googlebrotliprogrambench
https://programbench.com/task/wintermute-cell__ngrrram.8ea13c3/
wintermute-cell/ngrrram — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
wintermutecellprogrambench
https://programbench.com/task/drew-alleman__datasurgeon.d257cee/
Drew-Alleman/DataSurgeon — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
drewallemanprogrambench
https://programbench.com/task/crowdagger__crowbook.ea214d7/
crowdagger/crowbook — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
crowbookprogrambench
https://programbench.com/task/unhappychoice__gittype.34b72d0/
unhappychoice/gittype — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
programbench
https://programbench.com/task/antonmedv__walk.bf802ef/
antonmedv/walk — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
walkprogrambench
https://programbench.com/task/madler__pigz.fe4894f/
madler/pigz — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
pigzprogrambench
https://programbench.com/task/hairyhenderson__gomplate.05eb3aa/
hairyhenderson/gomplate — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
gomplateprogrambench
https://programbench.com/task/junegunn__fzf.b56d614/
junegunn/fzf — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
fzfprogrambench
https://programbench.com/task/svenstaro__genact.16f96e3/
svenstaro/genact — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
genactprogrambench
https://programbench.com/task/shashwatah__jot.a92aad8/
shashwatah/jot — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
jotprogrambench
https://programbench.com/task/danmar__cppcheck.0a5b103/
danmar/cppcheck — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
danmarcppcheckprogrambench
https://programbench.com/task/ducaale__xh.4a6e44f/
ducaale/xh — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
xhprogrambench
https://programbench.com/team/
Team — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
teamprogrambench
https://programbench.com/task/stranger6667__jsonschema.d52e881/
Stranger6667/jsonschema — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
jsonschemaprogrambench
https://programbench.com/task/jonas__tig.8334123/
jonas/tig — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
jonastigprogrambench
https://programbench.com/task/chmln__handlr.90e78ba/
chmln/handlr — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
programbench
https://programbench.com/task/chmln__sd.87d1ba5/
chmln/sd — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
sdprogrambench
https://programbench.com/task/byron__dua-cli.8570c15/
Byron/dua-cli — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
byronduacliprogrambench
https://arxiv.org/abs/2605.03546v1
[2605.03546v1] ProgramBench: Can Language Models Rebuild Programs From Scratch?
Abstract page for arXiv paper 2605.03546v1: ProgramBench: Can Language Models Rebuild Programs From Scratch?
language modelsfrom scratchprogrambenchrebuildprograms
https://programbench.com/task/lfos__calcurse.49180d5/
lfos/calcurse — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
lfoscalcurseprogrambench
https://programbench.com/task/mibk__dupl.1bf052b/
mibk/dupl — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
mibkprogrambench
https://programbench.com/task/zk-org__zk.10d93d5/
zk-org/zk — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
zkprogrambench
https://programbench.com/task/yoav-lavi__melody.f4af9b4/
yoav-lavi/melody — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
yoavlavimelodyprogrambench
https://programbench.com/task/o2sh__onefetch.e5958ce/
o2sh/onefetch — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
onefetchprogrambench
https://programbench.com/task/stacked-git__stgit.430027d/
stacked-git/stgit — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
stackedgitprogrambench
https://programbench.com/task/segmentio__chamber.5f93f5f/
segmentio/chamber — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
chamberprogrambench
https://programbench.com/task/mkj__dropbear.75f699b/
mkj/dropbear — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
mkjdropbearprogrambench
https://programbench.com/task/robertdavidgraham__masscan.b99d433/
robertdavidgraham/masscan — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
masscanprogrambench
https://programbench.com/task/elkowar__pipr.fae0b17/
elkowar/pipr — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
piprprogrambench
https://programbench.com/task/dalance__amber.69a0f52/
dalance/amber — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
dalanceamberprogrambench
https://programbench.com/task/sirwart__ripsecrets.34c9e03/
sirwart/ripsecrets — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
programbench
https://programbench.com/task/zevv__duc.a58fa4e/
zevv/duc — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
ducprogrambench
https://programbench.com/
ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
programbench
https://programbench.com/task/rbakbashev__elfcat.52f8cc7/
rbakbashev/elfcat — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
elfcatprogrambench
https://programbench.com/task/sigoden__argc.04a08f1/
sigoden/argc — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
programbench
https://programbench.com/task/sharkdp__fd.40d8eb3/
sharkdp/fd — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
fdprogrambench
https://programbench.com/task/peco__peco.4e58dad/
peco/peco — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
pecoprogrambench
https://programbench.com/task/noborus__trdsql.d8c5ff6/
noborus/trdsql — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
programbench
https://programbench.com/task/brocode__fblog.3b54330/
brocode/fblog — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
fblogprogrambench
https://programbench.com/task/johnkerl__miller.8d85b46/
johnkerl/miller — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
millerprogrambench
https://programbench.com/task/git-bahn__git-graph.87b4473/
git-bahn/git-graph — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
gitbahngraphprogrambench
https://programbench.com/task/jrnxf__thokr.09375ef/
jrnxf/thokr — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
programbench
https://programbench.com/task/alexpovel__srgn.89f943b/
alexpovel/srgn — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
srgnprogrambench
https://programbench.com/task/lymphatus__caesium-clt.a529b2e/
Lymphatus/caesium-clt — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
caesiumcltprogrambench
https://programbench.com/task/nachoparker__dutree.44e877d/
nachoparker/dutree — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
dutreeprogrambench
https://programbench.com/extended/
Extended Results — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
extended resultsprogrambench
https://arxiv.org/abs/2605.03546
[2605.03546] ProgramBench: Can Language Models Rebuild Programs From Scratch?
Abstract page for arXiv paper 2605.03546: ProgramBench: Can Language Models Rebuild Programs From Scratch?
language modelsfrom scratchprogrambenchrebuildprograms
https://programbench.com/task/yaa110__nomino.f892499/
yaa110/nomino — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
programbench
https://programbench.com/task/sharkdp__bat.f822bd0/
sharkdp/bat — ProgramBench
ProgramBench evaluates whether language models can rebuild programs from scratch.
batprogrambench