impulse

Step 3.7 Flash

Open weights

Step 3.7 Flash is StepFun's latest high-efficiency multimodal Mixture-of-Experts model.

API id Step-3.7-FlashReleased 28 May 2026 (May 2026)
Input / 1M
$0.20
at DeepInfra
Output / 1M
$1.15
at DeepInfra
Context window
262K
tokens
Max output
262K
tokens per reply
Hosts
6
6 listings
Released
May 2026
28 May 2026

Pricing

List prices at DeepInfra, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$0.20$1.15
Cached input
80% off input
$0.04

What it costs a month

At DeepInfra list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$132
  • RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$99.00
  • Coding agent1,000 requests a day, 40K in / 2K out, cached$175
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$756
Use your own numbers

Where to run it

6 hosts serve Step 3.7 Flash. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
step-3-7-flash
$0.20 / $1.15
DeepInfraInference
Step-3.7-Flash
$0.20 / $1.15
GMI CloudInference
Step-3.7-Flash
$0.20 / $1.15
Novita AIInference
step-3.7-flash
$0.20 / $1.15
step-3.7-flash
$0.223 / $1.28
+11% vs headline
AIHubMixGateway
step-3.7-flash
$0.22 / $1.32
+13% vs headline

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
TextImages
Produces
Text
Features
Prompt caching
Open weights
License: Apache-2.0
Runs in about 483.3 GB of GPU memory

Cheaper alternatives

Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.

Same model, cheaper host

No host sells it for less than DeepInfra.

Smaller Independent models

No cheaper Independent model keeps its key capabilities.

Other frontier labs

  • Claude Haiku 5.5
    Anthropic · on Anthropic
    $0.10 / $0.50
    Has every key capability
    54% less
  • GPT-6 Luna
    OpenAI · on OpenAI
    $0.10 / $0.50
    Has every key capability
    54% less
  • GPT-4o Mini
    OpenAI · on OpenAI
    $0.15 / $0.60
    Has every key capability
    40% less

Open-weight picks

  • MiMo V2.6 Pro
    Xiaomi · on GMI Cloud
    $0.348 / $0.696
    Has every key capability
    1% less
  • MiniMax M3
    MiniMax · on EmpirioLabs AI
    $0.225 / $0.90
    Has every key capability
    10% less
  • Muse Glimmer 30B
    Meta · on EmpirioLabs AI
    $0.20 / $0.80
    Has every key capability
    20% less
  • Deepseekv4.1 Flash
    DeepSeek · on Sarvam
    $0.149 / $0.585
    Has every key capability
    41% less

Model line

Newer and older versions, at each lab's own price.

  1. May 2026Step 3.7 FlashLatest$0.20 / $1.15
  2. UndatedStep 3.5 Flash$0.10 / $0.30

Call it

Using DeepInfra's API.

Python
from litellm import completion

response = completion(
    model="deepinfra/Step-3.7-Flash",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does Step 3.7 Flash cost?

On DeepInfra, Step 3.7 Flash costs $0.20 per million input tokens and $1.15 per million output tokens. Cached input costs $0.04 per million tokens.

Which APIs offer Step 3.7 Flash?

Step 3.7 Flash runs on 6 APIs: EmpirioLabs AI, DeepInfra, GMI Cloud, Novita AI, Kyma API and AIHubMix.

What is Step 3.7 Flash's context window?

262K tokens, with up to 262K tokens in one reply.

Is Step 3.7 Flash being retired?

No retirement date is scheduled for it on any API we track.

Is there a newer version of Step 3.7 Flash?

No. Step 3.7 Flash is the newest model in its line.

What are cheaper alternatives to Step 3.7 Flash?

Models that cost less and keep most of what it can do: Claude Haiku 5.5 on Anthropic (54% less), GPT-6 Luna on OpenAI (54% less) and MiMo V2.6 Pro on GMI Cloud (1% less). Savings are on a mix of 3 input tokens to 1 output token.

When was Step 3.7 Flash released?

Independent released it on 28 May 2026.

What is Step 3.7 Flash's API model id?

Use Step-3.7-Flash with DeepInfra. Other hosts use their own ids: step-3-7-flash on EmpirioLabs AI, step-3.7-flash on Novita AI, step-3.7-flash on Kyma API, step-3.7-flash on AIHubMix.

Sources

Raw catalog entry for Step-3.7-Flash on DeepInfra
{
  "max_output_tokens": 262144,
  "max_input_tokens": 262144,
  "input_cost_per_token": 2e-7,
  "output_cost_per_token": 0.00000115,
  "cache_read_input_token_cost": 4e-8,
  "supports_prompt_caching": true,
  "litellm_provider": "deepinfra",
  "mode": "chat",
  "supports_function_calling": true,
  "supports_tool_choice": true,
  "supports_vision": true,
  "source": "https://deepinfra.com/pricing"
}