Benchtalks #2: @jyangballin
, creator of SWE-bench, fresh off ProgramBench (every frontier model: 0% at launch). Out soon with @vincentsunnchen
.
Benchtalks #2: SWE-bench creator jyangballin discusses ProgramBench and upcoming project
By
–

By
–

Benchtalks #2: @jyangballin
, creator of SWE-bench, fresh off ProgramBench (every frontier model: 0% at launch). Out soon with @vincentsunnchen
.