# Step 3.7 Flash

> A 1T parameter multimodal Mixture-of-Experts model optimized for long-horizon coding, agentic tool use, and image/video understanding. Served via NVIDIA NIM with fast inference and structured output support.

**ID**: `stepfun-ai/step-3.7-flash`  
**Creator**: stepfun-ai  
**Category**: multimodal  
**Context**: 262K tokens  
**Input**: $0.20/M  
**Output**: $1.15/M  
**Released**: 2026-06-04  
**Web page**: https://anyrouter.dev/model/stepfun-ai/step-3.7-flash

**Input modalities**: text, image  
**Output modalities**: text  

**Tokenizer**: StepFun

**Capabilities**: chat, streaming, vision, coding, function-calling

## Usage

**Endpoint**: `POST https://anyrouter.dev/api/v1/chat/completions`

```bash
curl https://anyrouter.dev/api/v1/chat/completions \
  -H "Authorization: Bearer $ANYROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "stepfun-ai/step-3.7-flash",
  "messages": [
    {
      "role": "user",
      "content": "Say hi in 3 words."
    }
  ]
}'
```

## Providers

| Provider | Input | Output |
| --- | --- | --- |
| kilo | $0 | $0 |
| openrouter-byok | $0 | $0 |
| nousresearch-byok | $0 | $0 |
| hermes-agent-byok | $0 | $0 |
| stepfun-byok | $0 | $0 |
| stepplan-byok | $0 | $0 |
| commandcode-byok | $0 | $0 |

## Benchmarks

- **GPQA Diamond**: 0.8
- **HLE**: 0.2
- **τ²-Bench Telecom**: 1.0
- **SciCode**: 0.4

## Source

- Upstream docs: https://build.nvidia.com/stepfun-ai/step-3.7-flash