Step 3.5 Flash

reasoningapache-2.0

Flagship reasoning model with 256k context, MoE architecture (199B total, ~11B active). Optimized for agentic tasks, tool calling, planning, math, and coding.

Context
256K
Input price
$0.1/1M tokens
Output price
$0.29/1M tokens

Capabilities

Reasoning & Interaction
Reasoning
Tool Calling
Output Control
Structured Output
JSON Mode
Infrastructure
Streaming
Fine-tuning

Details

Model IDstep-3.5-flash
Provider StepFun
Creator stepfun
Familystep-3.5
Licenseapache-2.0
Statusactive
Typereasoning
Input modalitiestext
Output modalitiestext
Open weightYes
Parameters—
Architecture—
Context window256K tokens
Max input—
Max output—
Reasoning tokensYes
Knowledge cutoff—
Release date—
Sourceofficial

Tools

Function Callingfunction_calling
Call external functions and APIs

Endpoints

Chat CompletionsPOST
Generate chat responses with messageshttps://api.stepfun.com/v1/chat/completions

Pricing

Input
$0.1
Output
$0.29
Cache write
—
Cache read
$0.02
Batch in
—
Batch out
—

Family Comparison: step-3.5

ModelContextMax outPricing
Step 3.5 Flash 2603256K—$0.1$0.29
Step 3.5 Flash256K—$0.1$0.29

API

GET/v1/models/stepfun/step-3.5-flash

Step 3.5 Flash

step-3.5-flash

Flagship reasoning model with 256k context, MoE architecture (199B total, ~11B active). Optimized for agentic tasks, tool calling, planning, math, and coding.

Changes · 4 entries
step-3.5-flashupdateca7a400Nov 2, 2026, 03:02 AM
capabilitiesfine tuningyes
tools["function_calling"]
step-3.5-flashupdateb2f45bfOct 29, 2026, 03:29 AM
statusdeprecated→active
step-3.5-flashupdate0ed066aJul 31, 2026, 11:38 AM
statusactive→deprecated
tools["function_calling"]
step-3.5-flashcreate3e56cf1Apr 26, 2026, 07:23 PM