1,565 emails, six decision models Jev project

1,565 emails, six decision models is a Benchmarks and evals project built with Jev. On Nikhil Mudholkar's 1,565-email test of six models, Perplexity v1.1 ranked

On Nikhil Mudholkar's 1,565-email test of six models, Perplexity v1.1 ranked highest, OpenAI was fastest, and Jev never missed when it claimed 99 percent certainty.

Category: Benchmarks and evals. Built with Jev.

Author: Nikhil Mudholkar

Official source

More Jev projects are on Jev projects and apps. Related: Laya projects and apps.

Similar projects in Benchmarks and evals

• A browser step from 2 s to 0.2 s
• A correction about generating haiku with Jev
• A decision brain, not an assistant
• A faster, cheaper path for business models
• A free tool that scores how rigorous a paper is
• A Jev idea that hit a wall
• A Jev speed test app, made public
• A number two open-weight Jev-like model