SWE Decision Index v0.3 Adds Private Benchmarks and Vision to Curb Gaming
multimodalart released SWE Decision Index v0.3, which tracks model decision-making with a stronger benchmark suite, a private half-weight test set to prevent gaming, and newly added vision evaluations.
2026-10-07 ~ 2026-10-07 · 4 related posts
- jev Decision Index v0.3 arrives with private benchmark half and vision rankings — multimodalart · 2026-10-07
- SWE Decision Index v0.3 adds private benchmarks and vision evaluation — multimodalart · 2026-10-07
- jev Decision Index v0.3 adds stronger benchmarks, private half and vision — mervenoyann · 2026-10-07
- jev Decision Index v0.3 adds a private benchmark half and vision tasks to curb benchmark gaming — maximelabonne · 2026-10-07