Artificial Analysis Launches Optima for Custom Model Benchmarking

ArtificialAnlys · x · 2026-08-19

Artificial Analysis launched Optima, a platform for benchmarking models on your own workloads. It enables users to build custom benchmarks around specific tasks to compare performance, speed, and cost efficiency. The platform supports the full workflow of building, running, and grading benchmarks, with evaluation types including Q&A, document input, and agentic tasks. Users can draft tasks and rubrics via an agent or import data to find the best model for their specific use case.

Related event: Artificial Analysis Launches Optima for Custom Model Benchmarking(2 posts)→

Original post →

More from Apps

Apps channel →