StepFun: Step 3.5 Flash

StepFun flash lane for quick multimodal reasoning and coding assistance

Source-linked Open weights Released 2026-01-29

API record Report

InputT

OutputT

Input price$0.1/M

Output price$0.3/M

Context256K

Max output256K

Providers14

Inference availability

Providers

Provider-specific identifiers, limits, and listed prices per million tokens. Every row links back to the provider's own documentation.

Report a price

Providers offering Step 3.5 Flash
Provider	Provider model ID	Context	Max output	Input	Output	Cache read	Capabilities	Docs
Nvidia	`stepfun-ai/step-3.5-flash`	256K	16.384K	—	—	—	ReasoningTools	Docs ↗
StepFun Step Plan (Global)	`step-3.5-flash`	256K	256K	—	—	—	ReasoningTools	Docs ↗
Qiniu	`stepfun/step-3.5-flash`	64K	4.096K	—	—	—	Not listed	Docs ↗
StepFun Step Plan (China)	`step-3.5-flash`	256K	256K	—	—	—	ReasoningTools	Docs ↗
Eden AI	`deepinfra/stepfun-ai/Step-3.5-Flash`	262.144K	256K	$0.09	$0.3	$0.02	ReasoningToolsJSON	Docs ↗
Vercel AI Gateway	`stepfun/step-3.5-flash`	262.114K	262.114K	$0.09	$0.3	$0.02	ReasoningTools	Docs ↗
StepFun (China)	`step-3.5-flash`	256K	256K	$0.1	$0.3	$0.02	ReasoningTools	Docs ↗
ZenMux	`stepfun/step-3.5-flash`	256K	64K	$0.1	$0.3	—	Tools	Docs ↗
Kilo Gateway	`stepfun/step-3.5-flash`	262.144K	65.536K	$0.1	$0.3	—	ReasoningTools	Docs ↗
OpenRouter	`stepfun/step-3.5-flash`	262.144K	65.536K	$0.1	$0.3	—	ReasoningTools	Docs ↗
Hugging Face	`stepfun-ai/Step-3.5-Flash`	262.144K	256K	$0.1	$0.3	—	ReasoningTools	Docs ↗
StepFun (Global)	`step-3.5-flash`	256K	256K	$0.1	$0.3	$0.02	ReasoningTools	Docs ↗
NanoGPT	`stepfun-ai/step-3.5-flash`	262.144K	65.536K	$0.1	$0.3	$0.05	Reasoning	Docs ↗
EmpirioLabs AI	`step-3-5-flash`	256K	131.072K	$0.1	$0.3	$0.02	ReasoningToolsJSON	Docs ↗

Capability badges appear only where the provider catalog explicitly lists support. A blank cell means the source is silent, not that the feature is absent.

Listed rates

Price across providers

Input price per million tokens as published by each provider. Bars are drawn from listed rates only — no traffic weighting, since the catalog observes no requests.

Lowest input $0.09/M

Across 10 priced providers

Median input $0.1/M

Midpoint of listed rates

Highest input $0.1/M

1.1× the lowest listed rate

Output range $0.3 – $0.3

Per million output tokens

Eden AI $0.09/M

Vercel AI Gateway $0.09/M

StepFun (China) $0.1/M

ZenMux $0.1/M

Kilo Gateway $0.1/M

OpenRouter $0.1/M

Hugging Face $0.1/M

StepFun (Global) $0.1/M

NanoGPT $0.1/M

EmpirioLabs AI $0.1/M

Cost calculator

Estimate a workload

Provider Input tokens 1,000,000 Output tokens 250,000 Cached input 0%

$0.00

Excludes taxes, non-token charges, and tiered discounts.

2 providers list the identical $0.09 input rate, so price alone will not separate them — compare context limits, max output, and capabilities above.

Context limits also differ by provider, from 64K to 262.144K tokens. Compare the provider table above before choosing on price alone.

Specification

Capabilities

Recorded from the source catalog and provider listings.

× Reasoning No

× Tool calling No

× Structured output No

× Attachments No

× Vision input No

✓ Open weights Yes

Creator: StepFun
Model family: Not documented
Knowledge cutoff: 2025-01
License: Not documented
Release date: 2026-01-29
Model ID: stepfun/step-3.5-flash

Weights: Hugging Face ↗

Published evaluations

Benchmarks

Every result stays attached to its source, version, metric and harness. Scores from different versions are never merged, and the profile below plots each benchmark against its own population rather than on a shared scale.

Benchmark registry

Benchmark profile

Each axis is this model's percentile among the 4 benchmarks it has published results for, measured against every other model with a score on that same benchmark. Percentiles are used because benchmarks do not share a scale — a 60 on one is not a 60 on another. Hover any point for the raw score.

SWE-Bench Verified resolved Source ↗

SWE-Bench Verified #17 of 32

74.4

SciCode percent correct · 2026-06-02 Source ↗

SciCode #3 of 27

40.4

Artificial Analysis Coding Index index · 2026-06-02 Source ↗

Artificial Analysis Coding Index #5 of 26

31.6

Terminal-Bench Hard success rate · 2026-06-02 Source ↗

Terminal-Bench Hard #4 of 22

27.3

Each strip shows every published score for that benchmark, with this model marked. The lighter shape behind the ticks is the density of results, and the dashed line is the median.

Cost against capability

Price and performance

Listed input price plotted against SWE-Bench Verified, the benchmark with the widest published coverage that this model appears in.

The stepped line is the efficient frontier: at each price, the best score available for that money or less. A model sitting on it is not being beaten by anything cheaper. Price is log-scaled because listed rates span four orders of magnitude. Only models with both a listed price and a score on this benchmark can appear.

Catalog activity

Change log

Field-level changes detected between successful source imports.

Full change log

Sep 11, 2026Max Output Tokens65536 → 256000

Sep 11, 2026Context Length262144 → 256000

Sep 11, 2026Price Prompt0.09999999999999999 → 0.1

Sep 11, 2026Max Output Tokens256000 → 65536

Sep 11, 2026Context Length256000 → 262144

Sep 11, 2026Price Prompt0.1 → 0.09999999999999999

Sep 11, 2026Max Output Tokens65536 → 256000

Sep 11, 2026Context Length262144 → 256000

Sep 11, 2026Price Prompt0.09999999999999999 → 0.1

Sep 11, 2026Max Output Tokens256000 → 65536

Provenance

Sources & verification

Every figure on this page traces back to one of these records.

Methodology

Last verified: Sep 11, 2026
Status: Source-linked
Confidence: Medium
Catalog source: Models.dev

Public API

Use this record

Fetch the complete source-linked model record. No key, no account, no rate-limited tier.

API documentation

Endpoint

GET https://model.kyssta.lol/api/v1/models/stepfun/step-3.5-flash

curl

curl "https://model.kyssta.lol/api/v1/models/stepfun/step-3.5-flash"

Common questions

Frequently asked questions

Answered directly from the stored record — nothing here is generated beyond the catalog's own fields.

What is Step 3.5 Flash?

StepFun flash lane for quick multimodal reasoning and coding assistance. It is published by StepFun and catalogued here from Models.dev.

How much does Step 3.5 Flash cost?

Listed input pricing starts at $0.09 per million tokens from Eden AI, rising to $0.1 across 10 listed providers.

What is the context length of Step 3.5 Flash?

Step 3.5 Flash accepts up to 256K tokens of context and returns up to 256K output tokens.

Which providers serve Step 3.5 Flash?

14 providers list this model: Nvidia, StepFun Step Plan (Global), Qiniu, StepFun Step Plan (China), Eden AI, Vercel AI Gateway and 8 more.

Are the weights for Step 3.5 Flash open?

Yes. The weights are published and downloadable from Hugging Face.

When was Step 3.5 Flash released?

The catalog records a release date of 2026-01-29, last verified Sep 11, 2026.

More models from StepFun

View all →

Step 3.7 Flash

Newer StepFun flash model for faster agents, coding, and multimodal prompts

256K ctx $0.185/M in Step 3.5 Flash 2603

StepFun flash model for efficient multimodal reasoning, coding, and tool use

256K ctx $0.1/M in