StepFun (China) — kaged models

id: stepfun

npm: @ai-sdk/openai-compatible

env: STEPFUN_API_KEY

api: https://api.stepfun.com/v1

doc: https://platform.stepfun.com/docs/zh/overview/concept

Models

Step 1 (32K)

step-1-32k

in $2.05/M

out $9.59/M

cache read $0.41/M

ctx: 32,768 max out: 32,768 in: text out: text

reasoning tools vision structured temp open weights

Step 2 (16K)

step-2-16k

in $5.21/M

out $16.44/M

cache read $1.04/M

ctx: 16,384 max out: 8,192 in: text out: text

reasoning tools vision structured temp open weights

Step 3.5 Flash

step-3.5-flash

in $0.10/M

out $0.30/M

cache read $0.02/M

ctx: 256,000 max out: 256,000 in: text out: text

reasoning tools vision structured temp open weights

Step 3.5 Flash 2603

step-3.5-flash-2603

in $0.10/M

out $0.30/M

cache read $0.02/M

ctx: 256,000 max out: 256,000 in: text out: text

reasoning tools vision structured temp open weights

Step 3.7 Flash

step-3.7-flash

in $0.18/M

out $1.11/M

cache read $0.04/M

ctx: 256,000 max out: 256,000 in: text, image, video out: text

reasoning tools vision structured temp open weights

Step TTS 2

step-tts-2

in —

out —

ctx: 0 max out: 0 in: text out: audio

reasoning tools vision structured temp open weights

StepAudio 2.5 ASR

stepaudio-2.5-asr

in —

out —

ctx: 0 max out: 0 in: audio out: text

reasoning tools vision structured temp open weights

StepAudio 2.5 TTS

stepaudio-2.5-tts

in —

out —

ctx: 0 max out: 0 in: text out: audio

reasoning tools vision structured temp open weights