LodeHQ
AI news for people who shipFree, by email

AI · arxiv.org · Thursday, 5 November 2026

ViBench 2.0 tests whether coding agents can finish products

ViBench 2.0 gives developers a tougher measure of agent reliability: whether an agent can turn 241 successive feature requests into a working product, not merely solve one coding task. Its browser agent runs 544 test plans across 17 full-stack products, and a build passes only when every plan passes. The strongest three models pass up to 98% of plans but finish a fully working product in no more than 65% of builds; no open-weight model does so.

Read the source: arxiv.org ↗In the 5 November issue →