Benchmarks
Benchmarks
FrontierSWE v2
Codingunit % · normalized over [0, 80]Proximal's FrontierSWE v2 long-horizon software-engineering benchmark (distinct from FrontierCode).
#ModelSourceScoreNormalized
Score distribution
3 tracked results across the normalization window
080
Score vs. parameters
Open-weights models, log-x params
No open-weights models with disclosed parameter counts have a score here yet.