Together AI· Inference platformStory of the weekDeveloper tool

Together Evaluations: Benchmark Models for Your Tasks

Together AI announces an early preview of Together Evaluations, a platform for benchmarking LLM response quality using LLM-as-a-judge with three modes: classify, score, and compare.

From the source

we are releasing an early preview of Together Evaluations — a fast, flexible way to benchmark LLM response quality using leading open-source judge models you control.

together.ai

Announced