> ## Documentation Index
> Fetch the complete documentation index at: https://docs.kymaapi.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Step 3.7 Flash

> StepFun's fast flash tier — 256K context, multimodal input, tool calling, and cheap throughput.

## Overview

`step-3.7-flash` is StepFun's fast flash model and the successor to Step 3.5 Flash. It adds tool calling and runs a 256K context window with multimodal input (text, image, video) at a cheap cost band. Reach for it when you want fast, low-cost throughput with tool support and the option of image or video input.

## Specs

| Field              | Value                                               |
| ------------------ | --------------------------------------------------- |
| Model ID           | `step-3.7-flash`                                    |
| Best for           | Cheap fast throughput, multimodal input, bulk tasks |
| Context window     | 256K                                                |
| Max output tokens  | 8K                                                  |
| Input modalities   | Text, Image, Video                                  |
| Output modalities  | Text                                                |
| Tool calling       | Yes                                                 |
| Structured outputs | Yes                                                 |
| Reasoning          | Yes                                                 |
| Prompt caching     | Yes                                                 |
| Speed              | Fast                                                |
| Cost band          | Cheap                                               |
| Release stage      | Stable                                              |

## Use this when

* You want cheap, fast throughput on bulk or high-volume tasks.
* You need tool calling on a low-cost model.
* You want image or video input without paying a premium tier.

## Pick something else when

* You want the highest general quality: use [`qwen-3.7-plus`](/models/qwen-3.7-plus).
* You want a strong agentic coding model: use [`minimax-m3`](/models/minimax-m3).
* You want top deep reasoning: use [`deepseek-r1`](/models/deepseek-r1).

## Example

```python theme={null}
from openai import OpenAI

client = OpenAI(base_url="https://kymaapi.com/v1", api_key="ky-...")

response = client.chat.completions.create(
    model="step-3.7-flash",
    messages=[{"role": "user", "content": "Summarize this transcript into 5 bullet points."}]
)
```

## Agent query example

```bash theme={null}
curl "https://kymaapi.com/v1/models?recommended_for=cheap&tools=true"
```