Vals AI News
Follow us on
Follow us on

We're bringing independent AI evaluation to government.


Excel Modeling Benchmark Released
Vals AIAnthropic's Claude Sonnet 5 evaluated on the Vals Index
Vals AICyberBench Released
Vals AILegal Research Bench Released
Vals AIHarvey's Legal Agent Benchmark Released
Vals AIz.AI's GLM 5.2 evaluated across our benchmark suite
Vals AICode Migration Released
Vals AIKimi K2.7 Code evaluated across our coding benchmarks
Vals AIAnthropic’s safety warnings may have just backfired — the government has pulled the plug on its most powerful AI
TechCrunchAnthropic's Claude Fable 5 evaluated across our benchmark suite
Vals AI