to do
Benchmark Elastic agents
Nobody knows which models are actually usable in Elastic Agent Builder. Comprehensive benchmark across models, publish results.
hero · agent builder eval
date
— · 2026
status
to do
repo
llermaly/agent-builder-benchmarks
the problem →
Nobody knows which models are actually usable in Elastic Agent Builder.
what I built →
Comprehensive benchmark across models, publish results.
The story
Write the chapter for Agent Builder Eval here.
keep exploring →
Other chapters in this notebook