AI product
Vals AI is an AI benchmarking and evaluation company that independently tests models on economically valuable real-world tasks, including software, finance, law, cybersecurity and other frontier-risk areas. It develops and runs its own evaluations, publishes benchmark results and model comparisons through the Vals Index and specialized benchmarks such as finance-agent, Excel-modeling, tax-agent and legal evaluations, and reports performance alongside cost, latency and token usage. The videos also describe Val Smith, an internal-tooling product that evaluates models against a firm's actual GitHub work to help choose models and reduce token spending.
1 use taken from transcripts — each links to the moment in the video.
Builds AI benchmarks and evaluations, including the Vals Index and finance-focused benchmarks (finance agent, Excel modeling, tax agent) that frontier model labs cite in their model cards. Its evals measure the cost/performance Pareto frontier of models on real tasks and, based on internal tooling, it exposes a product called Val Smith that evaluates which models can do a firm's actual GitHub work, helping cut token spend.
1 in the library.