Step 3.5 Flash 2603
by StepFun Step Plan (China) · listed Apr 2026RTW
StepFun flash model for efficient multimodal reasoning, coding, and tool use
Current rates USD per 1M tokens
Input—
Output—
Cache read—
Cache write—
Against the market
On the record
- Model id
step-3.5-flash-2603- Context window
- 256K tokens
- Max output
- 256K tokens
- Input modalities
- text
- Output modalities
- text
- Reasoning
- yes
- Tool calls
- yes
- Open weights
- yes
- Knowledge cutoff
- Jan 2025
- Released
- 2026-04-02
- Last updated
- 2026-04-02
Also listed at
- StepFun (Global)$0.10 in · $0.30 out
- StepFun (China)$0.10 in · $0.30 out
- NanoGPT$0.10 in · $0.30 out
- StepFun Step Plan (Global)— in · — out