---
title: "Step 5 Preview AI model: facts and benchmarks | Kendr"
canonical: "https://kendr.org/models/stepfun-step-5-preview"
date_modified: "2026-10-13"
profile_kind: "reference"
---

# Step 5 Preview model profile

Step 5 Preview is a StepFun model profile with a 1,000,000-token context window. The dated catalog snapshot records text, image, video modalities and interfaces for prompt caching, reasoning controls, structured output, text generation, tool calling.

Step 5 Preview is StepFun's flagship model for agentic work, built on a sparse Mixture-of-Experts architecture (27B active / 600B total parameters). It performs strongly in software engineering and professional knowledge work, with particular strength in finance. It is designed for extended tasks that span large codebases and documents, using tools and refining results over multiple steps.

Step 5 Preview is a reference profile: StepFun publishes it, Kendr documents it for comparison, and it carries no Kendr alias to call.

Step 5 Preview is an independent knowledge profile. It does not claim that Kendr hosts the model, quote a Kendr customer price, promise API availability, or advertise a Kendr routing receipt.

Step 5 Preview accepts text, image, and video and returns text. The snapshot records prompt caching, reasoning controls, structured output, text generation, tool calling, video input, and vision as capabilities or interfaces.

The dated reference snapshot lists $1 input and $2.7 output per million tokens. Cached input is $0.05 per million tokens.

No comparable third-party benchmark value is present in the 2026-10-13 snapshot. Missing values are not estimated.

Step 5 Preview ranked #3 in the cited trailing-7-day third-party catalog usage dataset as of 2026-10-13. Observed within the cited trailing-7-day third-party catalog usage dataset; this is not global AI market share.

Profile data was reviewed for the 2026-10-13 snapshot. Claims retain their source dates, and unavailable fields are shown as unavailable instead of estimated.

## Profile facts

- **Profile type:** Model research profile
- **Provider or publisher:** StepFun
- **Context window:** 1,000,000 tokens
- **Snapshot date:** 2026-10-13
- **Knowledge cutoff:** Not disclosed
- **Reference catalog ID:** stepfun/step-5-preview

## When to pick Step 5 Preview

Step 5 Preview is a reference profile: StepFun publishes it, Kendr documents it for comparison, and it carries no Kendr alias to call.

Each condition below is derived from this model's own values in the 2026-10-13 snapshot, compared only against models publishing the same field.

- Reach for it when the input includes images. The snapshot records image input, so screenshots, scans, and diagrams can be sent directly instead of described.
- Reach for it when the model has to call tools. Tool calling is recorded in the snapshot, so this route can drive an agent loop rather than only answer in prose.
- Look elsewhere when you need to call it through Kendr today. This is a research reference profile with no Kendr alias. The facts and evidence here are published for comparison; the model is not routable on Kendr.
- Look elsewhere when the request is trivial and latency-sensitive. Reasoning is mandatory on this route and cannot be turned off, so short factual requests still pay reasoning tokens and reasoning latency.
- Look elsewhere when the workload is production traffic. The catalog marks this model preview. Preview routes can change behaviour or availability without the notice a stable route carries.

## Context, modalities, and identity

Step 5 Preview accepts text, image, and video and returns text. The snapshot records prompt caching, reasoning controls, structured output, text generation, tool calling, video input, and vision as capabilities or interfaces.

- **Provider or publisher:** StepFun
- **Context window:** 1,000,000 tokens
- **Maximum output:** 64,000 tokens
- **Input modalities:** text, image, and video
- **Output modalities:** text
- **Knowledge cutoff:** Not disclosed

## Model overview

Step 5 Preview is StepFun's flagship model for agentic work, built on a sparse Mixture-of-Experts architecture (27B active / 600B total parameters). It performs strongly in software engineering and professional knowledge work, with particular strength in finance. It is designed for extended tasks that span large codebases and documents, using tools and refining results over multiple steps.

- **Reference model ID:** stepfun/step-5-preview
- **Canonical version:** stepfun/step-5-preview-20261008
- **Hugging Face ID:** Not separately documented

## Capabilities and supported controls

Capabilities and parameters are reported from the dated catalog and source records; their presence does not guarantee identical behavior across every provider route.

- **Supported parameters:** frequency_penalty, include_reasoning, logprobs, max_tokens, reasoning, reasoning_effort, response_format, stop, structured_outputs, temperature, tools, top_logprobs, top_p

- prompt caching
- reasoning controls
- structured output
- text generation
- tool calling
- video input
- vision

## Dated pricing snapshot

The dated reference snapshot lists $1 input and $2.7 output per million tokens. Cached input is $0.05 per million tokens.

These are dated third-party reference prices, not Kendr prices or an availability offer. Routes, tiers, caching, region, and provider policy can change the landed price.

- **Price date:** 2026-10-13
- **Input:** $1 per 1M tokens
- **Cached input:** $0.05 per 1M tokens
- **Output:** $2.7 per 1M tokens

## Provider routes, performance, and uptime

The snapshot retains 1 provider endpoint. Provider prices, context limits, p50 performance, and uptime can differ by route and are not Kendr guarantees.

- **Active provider endpoints:** 1
- **Best p50 latency:** 3.45 s
- **Best p50 throughput:** 80 tok/s
- **Availability with routing:** 99.25% over the sampled window
- **Availability without routing:** 81.02% over the sampled window
- **Performance date:** 2026-10-13

| Provider | Quantization | Input / 1M | Output / 1M | Cache read / 1M | Context | p50 latency | p50 throughput | Uptime (1d) |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| StepFun | fp8 | $1 / 1M | $2.7 / 1M | $0.05 / 1M | 1,000,000 tokens | 3.45 s | 80 tok/s | 100.00% |

## Benchmark evidence and limitations

No comparable third-party benchmark value is present in the 2026-10-13 snapshot. Missing values are not estimated.

Third-party catalog fields normalized from the cited model dataset; benchmark methodology and coverage differ by source.

## Popularity and market context

Step 5 Preview ranked #3 in the cited trailing-7-day third-party catalog usage dataset as of 2026-10-13. Observed within the cited trailing-7-day third-party catalog usage dataset; this is not global AI market share.

- **Third-party catalog rank:** #3
- **Global market share:** Not inferred
- **Observation date:** 2026-10-13

## Frequently asked questions

### When should I use Step 5 Preview?

Step 5 Preview is a reference profile: StepFun publishes it, Kendr documents it for comparison, and it carries no Kendr alias to call. Reach for it when the input includes images: The snapshot records image input, so screenshots, scans, and diagrams can be sent directly instead of described. Reach for it when the model has to call tools: Tool calling is recorded in the snapshot, so this route can drive an agent loop rather than only answer in prose. Look elsewhere when you need to call it through Kendr today: This is a research reference profile with no Kendr alias. The facts and evidence here are published for comparison; the model is not routable on Kendr. Look elsewhere when the request is trivial and latency-sensitive: Reasoning is mandatory on this route and cannot be turned off, so short factual requests still pay reasoning tokens and reasoning latency. Look elsewhere when the workload is production traffic: The catalog marks this model preview. Preview routes can change behaviour or availability without the notice a stable route carries.

### What is Step 5 Preview?

Step 5 Preview is a StepFun model profile with a 1,000,000-token context window. The dated catalog snapshot records text, image, video modalities and interfaces for prompt caching, reasoning controls, structured output, text generation, tool calling.

### What context window does Step 5 Preview have?

The dated profile lists 1,000,000 tokens of context and up to 64,000 output tokens. Provider routes, variants, and runtime configuration can impose lower effective limits.

### What benchmark evidence is available for Step 5 Preview?

No comparable third-party benchmark value is present in the 2026-10-13 snapshot. Missing values are not estimated. Third-party catalog fields normalized from the cited model dataset; benchmark methodology and coverage differ by source.

### Is Step 5 Preview available through Kendr?

This is an independent knowledge profile, not a Kendr-hosted availability claim. Check Kendr's live public model API for currently enabled Kendr aliases.

### Does the popularity rank represent Step 5 Preview's global market share?

Step 5 Preview ranked #3 in the cited trailing-7-day third-party catalog usage dataset as of 2026-10-13. Observed within the cited trailing-7-day third-party catalog usage dataset; this is not global AI market share. No global market-share percentage is inferred when the source does not publish one.

## Sources and evidence dates

- [Step 5 Preview third-party catalog record](https://openrouter.ai/stepfun/step-5-preview) — catalog, checked 2026-10-13
- [Third-party model catalog methodology](https://openrouter.ai/docs/guides/overview/models) — methodology, checked 2026-10-13

Live Kendr operational availability is published separately at https://api.kendr.org/api/public/models.