Step 5 Preview

StepFun · released Oct 8, 2026

Open weightsReasoningTool useStructured outputVisionVideo in

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...

Compare this model Source listing

Specification

Context window: 1.05M
Max output: 64K
Knowledge cutoff: Not stated
Parameters: 604.3B
Licence: other
Serving providers: 2
Moderated: No
Uptime: 100.0%

Intelligence

43.7

94th percentile

Coding

—

Coding Index

Agentic

—

Agentic Index

Output speed

49 t/s

Median across providers

Latency

2.57s

Time to first token

Cost per task

$0.27

Estimated

Benchmarks

Where the score comes from

The Intelligence Index is a composite. These are the underlying evaluations this model was actually measured on.

Evaluation scores

Percentage correct · higher is better

AA-LCR (long context)
88.3%
SciCode
58.9%
Humanity's Last Exam
46.5%

An evaluation missing from this list was not run for this model — it is not a zero.

View as table

Evaluation	Score
AA-LCR (long context)	88.3%
SciCode	58.9%
Humanity's Last Exam	46.5%

Against its peers

Intelligence Index · this model highlighted, nearest peers in grey

Qwen3.8 Max (0902)Qwen
45.4
GLM 5.3Z.ai
44.8
Grok 4.6xAI
44.3
Step 5 PreviewStepFun
43.7
Kimi K3Moonshot AI
43.6
Claude Haiku 5.5Anthropic
43.4
GPT-5.6 TerraOpenAI
42.1
GLM 5.3 FlashZ.ai
41.8

Peers are the models sitting closest on the Intelligence Index — the set you would realistically choose between.

View as table

Model	Intelligence
Qwen3.8 Max (0902)	45.4
GLM 5.3	44.8
Grok 4.6	44.3
Step 5 Preview	43.7
Kimi K3	43.6
Claude Haiku 5.5	43.4
GPT-5.6 Terra	42.1
GLM 5.3 Flash	41.8

Percentile among all indexed models

Intelligence94th

Pricing

What it costs to run

List prices per million tokens, plus what one representative task works out to.

List price

Input / 1M tokens: $1
Output / 1M tokens: $2.7
Cached input / 1M: $0.05
Blended 3:1: $1.43

One task, estimated

$0.27

Input tokens: 50,000
Output tokens: 80,000
Profile: Reasoning

Estimated from list pricing: 50K input tokens plus 80K output tokens for reasoning models (25K for non-reasoning).

Serving providers

Speed and latency figures are medians across these providers, so a widely-served model reports a blend rather than any single endpoint.