modelbenchmark.io

Auto Value Reasoning

auto-value-reasoning

Compare

Routes to the model with the most reasoning points per measured dollar, among models within 5 points of the best

Specification

most-agreed values

Context
262K
Max output
64K
Released
2026-09-25
Knowledge cutoff
—
Retires
—
Open weights
no
Input
text
Output
text

Available from 1 host

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
aistupidlevel————262K64K—