Hermes Bench compares models on mean score and cost per task
Original titleEvery model runs the same harness, with reasoning set high where offered. The index reports the mean score and the mean cost per task acr...
AISummary
Nous Research's Hermes Bench runs every model through the same harness, with reasoning set high where offered. The index reports each model's mean score and mean cost per task across four suites.
Source: Nous Research · x.comPublished · added here