> Step 3.5 Flash is an open-weights AI model from StepFun, released 2 Feb 2026. API price $0.10 in / $0.30 out per million tokens, 262K context.

[Models](https://themodelindex.org/models/) /[StepFun](https://themodelindex.org/labs/stepfun/)

# Step 3.5 Flash

StepFun

Released 2 Feb 2026

Open

Date verified

[Compare](https://themodelindex.org/compare/step-3-5-flash-vs-gpt-5-3-codex/) [Announcement](https://huggingface.co/stepfun-ai/Step-3.5-Flash)

–

Index score

Not yet evaluated independently

–

Core tests taken

no independent results yet

–

Versus the frontier at release

Replaced by [Step 3.7 Flash](https://themodelindex.org/models/step-3-7-flash/)

$0.10/ $0.30

Price per million tokens

input / output, list price

262K

Context window

tokens it reads at once

## Where it sits · on the frontier

_Chart: Model Index score of every scored model since late 2022, with the best closed and best open model over time_

Not plotted: no index score yet.

## Test results · by area · tick = best by any model

Reasoning & knowledge

no evidence yet

GPQA Diamond

no result

Humanity's Last Exam

no result

Maths

no evidence yet

FrontierMath T1-3

no result

Coding

no evidence yet

SWE-bench Verified

no result

Terminal-Bench

no result

Agents

no evidence yet

APEX-Agents

no result

Novel problems

no evidence yet

ARC-AGI-2

no result

Long tasks

no evidence yet

METR time horizon

no result

Independent results from the Epoch AI benchmarking hub, best reasoning setting. The core tests are shown here; anchor tests also inform the score and can be shown above.

## What others say · rank on each leaderboard

LMArena · Maths

maths questions in chat

#129

of 373 · 1407

LMArena · Coding

coding questions in chat

#137

of 383 · 1449

LMArena · Text

overall chat quality, judged by people

#150

of 388 · 1393

LMArena · Creative writing

creative writing

#162

of 386 · 1345

Further left is better. Leaderboards measure different things: LMArena is people voting blind between two answers, the others are test scores. Click any row for the source.

## Best at · top three among current models

Replaced by Step 3.7 Flash; it no longer leads any category among current models.

Where this model places in the top three of today's models, on category leaderboards and on our core tests.

## Where to use it · prices checked 1 Oct 2026

#### From the lab

StepFun platform

API

↗

#### API providers · 1 · per million tokens, in / out

S

SiliconFlow

$0.10 / $0.30 · fp8

cheapest

#### Run it yourself

🤗

Official weights

stepfun-ai/Step-3.5-Flash

↗

Provider prices come from OpenRouter's live listing; going direct to a provider can differ. Subscriptions are the lab's own apps.

## Facts and sources

Released

2 Feb 2026

Announcement

[huggingface.co ↗](https://huggingface.co/stepfun-ai/Step-3.5-Flash)

Release entry

Step 3.5 Flash

Scored as

–

Artificial Analysis index

–

Output speed

–

Hugging Face

[stepfun-ai/Step-3.5-Flash ↗](https://huggingface.co/stepfun-ai/Step-3.5-Flash)

## StepFun releases · around this one

Step 3

Open

25 JUL 2025

Step 3.5 Flash

Open

2 FEB 2026

Step 3.7 Flash

Open

29 MAY 2026

Step 5 Preview

Closed

18 SEP 2026

~175 est.

---
Source: https://themodelindex.org/models/step-3-5-flash/ · The Model Index · data as of 2026-10-01