modelbenchmark.io

Sakana Namazu

sakana-namazu

Compare

Japanese-specialized reasoning model based on Kimi K2.6 and tuned for Japanese language, culture, and business workflows

Specification

most-agreed values

Context
262K
Max output
66K
Released
2026-08-03
Knowledge cutoff
Retires
Open weights
no
Input
text, image, pdf
Output
text

Price

US dollars per million tokens · most-agreed

Input
$0.95
Output
$4.00
Cache read
$0.15

Available from 4 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
kilo · sakana$0.95$4.00$0.15262K66K
merge-gateway · sakana$0.95$4.00$0.15262K66K
openrouter · sakana$0.95$4.00$0.15262K66K
sakana$0.95$4.00$0.15262K66K

Other listings of this model

same model, different key