modelbenchmark.io

Step 3.7 Flash

step-3-7-flash

Compare

StepFun flash model for efficient multimodal reasoning, coding, and tool use

Specification

most-agreed values

Context
262K
Max output
262K
Released
2026-05-29
Knowledge cutoff
2026-03-01
Retires
Open weights
yes
Input
text, image, video
Output
text

Price

US dollars per million tokens · most-agreed

Input
$0.19
Output
$1.14
Cache read
$0.03
Cache write
free

Available from 23 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
kenarifreefree256K256K
kenari · freefreefree256K256K
kilo · freefreefreefree262K262K
nvidia · stepfun-aifreefree256K16K
unorouter · freefreefree256K256K
stepfun$0.185$1.11$0.037256K256K
stepfun-ai$0.185$1.11$0.037256K256K
ambient · stepfun$0.19$1.14$0.03free262K262K
deepinfra$0.20$1.15$0.04262K
deepinfra · stepfun-ai$0.20$1.15$0.04262K256K
edenai$0.20$1.15$0.04262K256K
edenai · flexai$0.20$1.15262K256K
empiriolabs$0.20$1.15$0.04256K131K
huggingface · stepfun-ai$0.20$1.15262K256K
kilo · stepfun$0.20$1.15$0.04256K230K
nano-gpt · thinking$0.20$1.15$0.04262K256K
novita$0.20$1.15$0.04262K256K
openrouter · stepfun$0.20$1.15$0.04262K230K
requesty$0.20$1.15$0.04262K256K
vercel · stepfun$0.20$1.15$0.04256K256K
zenmux · stepfun$0.20$1.15256K256K
stepfun-ai-step-plan256K256K
stepfun-step-plan256K256K