jevPOST
Decision model evals: Jev leads on quality and latency
The author says their decision-model evals, based on scores in the Hugging Face Jev Decision Index, lead on the quality-latency Pareto frontier, and shares a live demo.
michelle
@michellechenArchived 2026.10.02Originally posted 2026.10.01
michelle @michellechenOriginally posted 2026.10.01
we currently lead the evals around decision models, pushing the pareto frontier on both quality and latency. these evals are based on the model scores in @huggingface Jev Decision Index. check out our live eval demo site here: https://clef-evals.workers-ai-mle.workers.dev/
x.com/michellechen/status/2105684871922270424Archived 2026.10.02 10:04