Hormouz AIHormouzSign in

StepFun: Step 3.7 Flash: AI model

Step 3.7 Flash is StepFun’s efficient multimodal model with native image and video understanding, a 256K context window and selectable reasoning levels. Reachable from inside Iran through Hormouz.

Context window:
262,144 tokens
Input:
$0.2/1M tokens
Output:
$1.15/1M tokens
Type:
text, image, video
  • Vision
  • Tools
  • Streaming
  • Long context

Thinking levels: Always thinks · high · medium · low

Step 3.7 Flash is a high-efficiency multimodal model built on a Mixture-of-Experts architecture. It pairs a 196B-parameter language backbone with a vision encoder and activates roughly 11B parameters per token, delivering native image and video understanding at a reasonable compute cost.

The model supports a context window of up to 256K tokens and exposes selectable reasoning levels (high/medium/low), so you can trade off speed, cost and depth of reasoning. Step 3.7 Flash is designed for coding, agentic workflows, structured outputs and long-context tasks, and is served through Hormouz behind an OpenAI-compatible API.

Pricing is usage-based: $0.20 per million input tokens and $1.15 per million output tokens. Hormouz provides stable access to the model from inside Iran.

Step 3.7 Flash is StepFun's latest high-efficiency multimodal Mixture-of-Experts model. It pairs a 196B-parameter language backbone with a vision encoder for native image and video understanding, activating roughly 11B parameters...

Quickstart

curl https://api.hormouz.net/v1/chat/completions \
  -H "Authorization: Bearer ng-YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "stepfun/step-3.7-flash",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Specs

VendorStepFun
TypeText, Image, Video
Context window262,144 tokens
Pricing$0.20/M in · $1.15/M out
APIChat Completions
CapabilitiesStreaming · Vision · Tools

Benchmarks

39.6
Coding Index
Better than 41% of models we compared

Reasoning

GPQA Diamond
Graduate-level scientific reasoning
80.9%
HLE
Humanity's Last Exam
21.4%
IFBench
Instruction-following
67.3%
AA-LCR
Long-context reasoning
73.7%
CritPt
Research-level physics reasoning
2.3%
AA Omniscience
Broad factual knowledge accuracy
25.8%
Non-Hallucination
Rate of avoiding fabricated answers
15%

Coding

SciCode
Python for scientific computing
43.9%
Terminal-Bench Hard
Agentic coding & terminal use
35.6%

Agentic

τ²-Bench
Tool-use agents in dual-control scenarios
98.5%
GDPval
Economically valuable tasks
18%

Source: Artificial Analysis

Design Arena

How Design Arena works ↗
3D
Top 55%
1,139Elo
Win rate: 40.8%Avg time: 104s
ASCII Art
Top 52%
1,161Elo
Win rate: 45%Avg time: 110s
Code
Top 43%
1,187Elo
Win rate: 42.7%Avg time: 116s
Data Viz
Top 52%
1,178Elo
Win rate: 42.9%Avg time: 87.8s
Game Dev
Top 51%
1,171Elo
Win rate: 39%Avg time: 149s
SVG
Top 73%
1,081Elo
Win rate: 37.3%Avg time: 63.3s
UI Component
Top 47%
1,184Elo
Win rate: 42.1%Avg time: 101s
Website
Top 41%
1,202Elo
Win rate: 44.3%Avg time: 117s

Metrics sourced from Design Arena

Code example

cURL
curl https://api.hormouz.net/v1/chat/completions \
  -H "Authorization: Bearer ng-YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "stepfun/step-3.7-flash",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'
Python
from openai import OpenAI
client = OpenAI(api_key="ng-YOUR_API_KEY", base_url="https://api.hormouz.net/v1")
response = client.chat.completions.create(
    model="stepfun/step-3.7-flash",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)
JavaScript
import OpenAI from "openai";
const client = new OpenAI({ apiKey: "ng-YOUR_API_KEY", baseURL: "https://api.hormouz.net/v1" });
const response = await client.chat.completions.create({
  model: "stepfun/step-3.7-flash",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);

Buy the StepFun: Step 3.7 Flash API, paid in rial

  • Buy credit with an Iranian bank card; no foreign card or account needed.
  • One Hormouz key reaches StepFun: Step 3.7 Flash and 600+ other models, and you pay only for what you use.
  • No VPN and no risk of a banned account; just change the base URL to api.hormouz.net/v1.
Create an account

Use via Hormouz

Connect to this model and 600+ others with one API key, then see usage down to each request.

Request company access

FAQ

How much does Step 3.7 Flash cost?
Billing is usage-based: $0.20 per million input tokens and $1.15 per million output tokens.
How do I access Step 3.7 Flash from inside Iran?
Create an account on Hormouz and generate an API key. Hormouz provides a stable path to the model from inside Iran, with no foreign credit card or circumvention tooling required.
Is Step 3.7 Flash compatible with OpenAI code?
Yes. Hormouz exposes an OpenAI-compatible API. Just change the base URL to Hormouz and swap in your Hormouz key. Your existing SDKs and code run without a rewrite.
Does Step 3.7 Flash handle image and video?
Yes. It is multimodal and offers native image and video understanding alongside text.
What are the reasoning levels in Step 3.7 Flash?
The model exposes selectable reasoning levels (high/medium/low) so you can balance speed, cost and depth. Its context window is up to 256K tokens.

Updated 2026-10-13