Home/Models
Models/StepFun (China)

Step 3.7 Flash

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 per token. The model supports a 256K context window and exposes selectable reasoning levels (high/medium/low), letting callers trade off speed, cost, and depth of reasoning.

Designed for coding, agentic workflows, structured outputs, and long-context productivity tasks.

stepfun/step-3.7-flash
Organization
StepFun (China)
Family
—
Providers
20
Context
256,000
Output limit
256,000
Knowledge
2026-03-01
Release
2026-05-29
Updated
2026-05-29
Weights
Open
Input
text, image, video
Output
text
Capabilities
Tools, Reasoning, Temperature

Providers

Compare every cataloged offer for this model, including its provider, route, token limits, price, and capabilities.

25 offers · USD per million tokens
ProviderModel IDContextOutputInput / output · 1MReasoningToolsStructuredDetails
AIHubMix
step-3.7-flash
256,000256,000$0.22 / $1.32YesYes—
View
Ambient
stepfun/step-3.7-flash
262,144262,144$0.19 / $1.14YesYesYes
View
Deep Infra
stepfun-ai/Step-3.7-Flash
262,144256,000$0.20 / $1.15YesYes—
View
DevPass (LLM Gateway)
step-3.7-flash
262,144256,000$0.20 / $1.15YesYes—
View
Eden AI
flexai/Step-3.7-Flash
262,144256,000$0.20 / $1.15YesYesYes
View
Eden AI
deepinfra/stepfun-ai/Step-3.7-Flash
262,144256,000$0.20 / $1.15YesYesYes
View
EmpirioLabs AI
step-3-7-flash
256,000131,072$0.20 / $1.15YesYesYes
View
Hugging Face
stepfun-ai/Step-3.7-Flash
262,144256,000$0.20 / $1.15YesYes—
View
Kenari
step-3-7-flash:free
256,000256,000$0.00 / $0.00YesYes—
View
Kenari
step-3-7-flash
256,000256,000$0.00 / $0.00YesYes—
View
Kilo Gateway
stepfun/step-3.7-flash:free
262,144262,144$0.00 / $0.00YesYesNo
View
Kilo Gateway
stepfun/step-3.7-flash
256,000230,400$0.20 / $1.15YesYesYes
View
LLM Gateway
novita/step-3.7-flash
262,144256,000$0.20 / $1.15YesYesNo
View
NanoGPT
stepfun/step-3.7-flash:thinking
262,144256,000$0.20 / $1.15YesYesYes
View
Novitavia OpenRouter
stepfun/step-3.7-flash
fp8		262,144256,000$0.20 / $1.15YesYesYes
View
StepFunvia OpenRouter
stepfun/step-3.7-flash
fp8		256,000230,400$0.20 / $1.15YesYesYes
View
Requesty
step-3.7-flash
262,144256,000$0.20 / $1.15YesYesYes
View
StepFun (China)
step-3.7-flash
256,000256,000$0.185 / $1.11YesYes—
View
StepFun (Global)
step-3.7-flash
256,000256,000$0.185 / $1.11YesYes—
View
StepFun Step Plan (China)
step-3.7-flash
256,000256,000— / —YesYes—
View
StepFun Step Plan (Global)
step-3.7-flash
256,000256,000— / —YesYes—
View
UnoRouter
step-3.7-flash:free
256,000256,000$0.00 / $0.00YesYes—
View
Vercel AI Gateway
stepfun/step-3.7-flash
256,000256,000$0.20 / $1.15YesYes—
View
ZenMux
stepfun/step-3.7-flash-free
256,000256,000$0.00 / $0.00YesYes—
View
ZenMux
stepfun/step-3.7-flash
256,000256,000$0.20 / $1.15YesYes—
View

Prices and limits apply to each offer. A dash means the value is not provided. OpenRouter offers show the hosting provider and routing channel separately.

Pricing

Compare token rates and cache charges across available routes and hosting configurations.

USD per million tokens
Input from$0.00Per 1M input tokens
Output from$0.00Per 1M output tokens
Available offers25Across 20 providers

Lowest listed input and output prices across the offers above; they may belong to different providers. These are base rates, before conditional pricing, discounts or additional fees.

Endpoint price comparisonLatest observed snapshot · 2 lowest-priced configurations
InputOutput
Novitanovita/fp8
$0.20$1.15
StepFunstepfun/fp8
$0.20$1.15
stepfun/step-3.7-flashObserved
Input
$0.20
Output
$1.15
Cache read
$0.04
Hosting prices through OpenRouter All endpoints and cache rates
EndpointInput / 1MOutput / 1MCache read / 1MCache write / 1M
Novitanovita/fp8$0.20$1.15$0.04—
StepFunstepfun/fp8$0.20$1.15$0.04—

Each row is a hosting configuration. Open an offer in Providers for its full pricing conditions and additional fees. Endpoint observations may differ from the routing catalog's base price.

Performance

Review reported latency and output-throughput percentiles from the latest observed rolling window.

Latest 30-minute observation

Time to first token · milliseconds

Latency distributionRolling 30-minute snapshot · 2 endpoints
P50P75P90P99
Novitanovita/fp8
4,234–10,698.84 ms
StepFunstepfun/fp8
5,657–19,098.67 ms
All latency percentiles 2 endpoints
EndpointP50P75P90P99
Novitanovita/fp84,234 ms6,265.5 ms8,309.7 ms10,698.84 ms
StepFunstepfun/fp85,657 ms7,780.25 ms10,497.9 ms19,098.67 ms

Output throughput · tokens per second

Throughput distributionRolling 30-minute snapshot · 2 endpoints
P50P75P90P99
Novitanovita/fp8
56–217.25 t/s
StepFunstepfun/fp8
46–231 t/s
All throughput percentiles 2 endpoints
EndpointP50P75P90P99
Novitanovita/fp856 t/s113 t/s153 t/s217.25 t/s
StepFunstepfun/fp846 t/s113 t/s167 t/s231 t/s

Percentiles describe the observed request distribution. These are snapshots of a rolling window, not a historical time series.

Uptime

Compare successful request rates across the latest 5-minute, 30-minute, and 24-hour observations.

Successful requests by window
Availability by endpointCurrent overlapping windows · 2 endpoints
Endpoint5 minutes30 minutes24 hours
Novitanovita/fp8
StepFunstepfun/fp8

Reported by OpenRouter at the time of collection; rate-limited requests are excluded. A dash means no measurement was reported. This is not a live availability check.

Benchmarks

Review attributed evaluation results on each source's original scale.

External evaluations
stepfun/step-3.7-flashObserved
Benchmark profileArtificial Analysis · shared source scale from 0 to 40
Coding39.6
Design ArenaCategoryEloWin rateRank
models3d1,14440.8%68
modelsasciiart1,16345.1%37
modelscodecategories1,18742.7%57
modelsdataviz1,17942.9%62
modelsgamedev1,17139%63
modelssvg1,08037.2%64
modelsuicomponent1,18542.3%59
modelswebsite1,20244.3%56

Sources: Artificial Analysis and Design Arena, via OpenRouter. Scores retain their original scales and are not OpenAI Suite ratings.

Detailed evaluations

Evaluation scoresAccuracy or item F1 · higher is better
tau bench verified airlineStepFun: Step 3.7 Flash
77.333%
gpqa diamondStepFun: Step 3.7 Flash
76.599%
EvaluationConfigurationSourceResultsDetails
models · 3dStep 3.7 Flashstepfun/step-3.7-flash-20260528Design Arena1,144 Elo
Details
models · uicomponentStep 3.7 Flashstepfun/step-3.7-flash-20260528Design Arena1,185 Elo
Details
models · datavizStep 3.7 Flashstepfun/step-3.7-flash-20260528Design Arena1,179 Elo
Details
Composite indicesStep 3.7 Flashstepfun/step-3.7-flash-20260528Artificial AnalysisIntelligence — · Coding 39.6 · Agentic —
Details
models · websiteStep 3.7 Flashstepfun/step-3.7-flash-20260528Design Arena1,202 Elo
Details
tau bench verified airlineStepFun: Step 3.7 Flashstepfun/step-3.7-flash-20260528OpenRouter77.333%
Details
models · codecategoriesStep 3.7 Flashstepfun/step-3.7-flash-20260528Design Arena1,187 Elo
Details
models · gamedevStep 3.7 Flashstepfun/step-3.7-flash-20260528Design Arena1,171 Elo
Details
models · svgStep 3.7 Flashstepfun/step-3.7-flash-20260528Design Arena1,080 Elo
Details
gpqa diamondStepFun: Step 3.7 Flashstepfun/step-3.7-flash-20260528OpenRouter76.599%
Details
models · asciiartStep 3.7 Flashstepfun/step-3.7-flash-20260528Design Arena1,163 Elo
Details

Sources: Artificial Analysis, Design Arena and OpenRouter. Evaluation configurations and scales differ; results are not interchangeable. Updated Oct 4, 2026.

Apps & session costs

Compare observed median session costs for applications using this model, grouped by conversation length.

30-day sample · through Sep 27, 2026
ApplicationModel routeTurnsMedian cost / sessionDetails
Hermes Agentstepfun/step-3.7-flash-202605282–9 turns
$0.008192
Details
Hermes Agentstepfun/step-3.7-flash-202605281 turn
$0.001317
Details
Hermes Agentstepfun/step-3.7-flash-2026052810–49 turns
$0.050307
Details

Source: OpenRouter, as of Oct 2, 2026. Published under CC BY 4.0. These are observed costs per session, not token prices or a forecast for your workload. Apps are compared separately.

Activity

Daily prompt and completion tokens reported for this model's routes in OpenRouter's top 50. Missing days have no published value.

Last 90 complete UTC days
Daily token usage9,024,067,047,469 reported tokens · 63 days with observations
Jul 6, 2026Oct 3, 2026
Daily observations · 63 days
Date (UTC)Reported tokensReported routes
Sep 7, 202650,383,448,1841
Sep 6, 202646,227,292,1441
Sep 5, 202643,822,821,8301
Sep 4, 202656,596,434,9691
Sep 3, 202673,630,465,7331
Sep 2, 202682,981,602,0561
Sep 1, 202683,766,919,9011
Aug 31, 202669,067,793,4941
Aug 30, 202654,418,413,7061
Aug 29, 202655,670,929,9951
Aug 28, 202661,199,591,2931
Aug 27, 202662,431,028,6511
Aug 26, 202664,428,228,7021
Aug 25, 202657,995,357,2631
Aug 24, 202662,292,417,2741
Aug 23, 202660,632,405,9951
Aug 22, 202655,823,852,2171
Aug 21, 202665,930,998,5931
Aug 20, 202669,043,568,8231
Aug 19, 202673,520,348,7281
Aug 18, 2026105,496,988,3641
Aug 17, 2026127,454,888,6251
Aug 16, 2026113,515,350,6031
Aug 15, 2026111,007,985,1091
Aug 14, 2026125,957,921,2571
Aug 13, 2026142,809,650,0191
Aug 12, 2026130,992,558,7801
Aug 11, 2026170,094,087,8521
Aug 10, 2026171,499,623,4231
Aug 9, 2026158,346,687,7621
Aug 8, 2026164,102,567,4521
Aug 7, 2026177,898,765,8241
Aug 6, 2026183,760,329,8481
Aug 5, 2026198,991,386,4791
Aug 4, 2026188,156,210,4111
Aug 3, 2026195,693,794,2171
Aug 2, 2026174,724,791,5181
Aug 1, 2026177,929,114,8761
Jul 31, 2026202,112,542,8791
Jul 30, 2026282,473,777,0781
Jul 29, 2026324,874,967,8631
Jul 28, 2026266,482,278,6341
Jul 27, 2026227,222,837,2351
Jul 26, 2026193,014,928,4321
Jul 25, 2026204,804,309,2441
Jul 24, 2026243,774,901,4731
Jul 23, 2026389,840,332,5961
Jul 22, 2026456,677,096,2251
Jul 21, 2026307,580,931,3261
Jul 20, 2026122,888,549,3231
Jul 19, 2026105,574,831,4021
Jul 18, 2026101,493,723,4991
Jul 17, 2026112,061,266,2161
Jul 16, 2026106,035,149,2841
Jul 15, 2026104,939,271,8121
Jul 14, 2026103,250,368,0931
Jul 13, 2026116,975,651,0801
Jul 12, 2026133,573,694,5491
Jul 11, 2026156,928,798,8171
Jul 10, 2026171,824,144,1151
Jul 9, 2026158,634,744,0021
Jul 8, 2026170,234,161,6081
Jul 7, 2026192,497,168,7141

Source: OpenRouter (openrouter.ai/rankings), as of Oct 4, 2026. CC BY 4.0. Totals include only individually published routes. Tokenizers differ by provider; the aggregated “other” category is never assigned to a model.

Technical details & API

Explore specifications, supported parameters, reasoning controls, and API identifiers for each OpenRouter route.

Current route specifications
stepfun/step-3.7-flashObserved
Context
262,144
Maximum output
230,400
Tokenizer
Other
Input
text, image, video
Output
text
Moderation
No
Added to OpenRouter
2026-05-28
Base URLhttps://openrouter.ai/api/v1
Model IDstepfun/step-3.7-flash
Version IDstepfun/step-3.7-flash-20260528
Weights IDstepfun-ai/Step-3.7-Flash

Reasoning

Required
Yes
Enabled by default
Not reported
Default effort
medium
Supported efforts
high, medium, low

Supported parameters

  • frequency_penalty
  • include_reasoning
  • logprobs
  • max_tokens
  • presence_penalty
  • reasoning
  • reasoning_effort
  • repetition_penalty
  • response_format
  • seed
  • stop
  • structured_outputs
  • temperature
  • tool_choice
  • tools
  • top_k
  • top_logprobs
  • top_p

Frequently asked questions

Answers based on the model metadata and serving offers currently in the catalog.

How much does Step 3.7 Flash cost?

Listed rates start at $0.00 per million input tokens and $0.00 per million output tokens. Prices vary by provider and configuration; see Pricing for details.

Which providers offer Step 3.7 Flash?

There are 25 cataloged offers across 20 providers. Compare model IDs, prices and limits in Providers.

What is the context window?

The cataloged model context is 256,000 tokens. Each serving endpoint may apply a different limit.

Does it support tools and structured output?

Tool calling: Yes. Structured output: Not reported. Support can vary by endpoint.

Resources

Hugging Face ↗
Catalog history · 2 recorded revisions

Metadata observations since this model was first cataloged. These are not model release versions.

  • · 8c9a397096a4
  • · f976fca56afd