How do semantic-search and Capability compare on challenges?

Choose a value for each blank above to run this.