TL;DR

Step 5 Preview by StepFun scores 69.72 overall, costs $1.00 per 1M input tokens and $2.70 per 1M output tokens, and carries a context window of 1M tokens — observed 2026-10-23.

Step 5 Preview

StepFun · Open Weight · rank 19 · bench-align-v5

Compare this model Model a workload

Canonical idstep-5-preview

Overall score69.72

Context window1M tokens

Release date2026-09-20

Access typeOpen Weight

Blended $/1M$1.4375% input / 25% output

Capability shape

Seven axes from the ranking source. A missing axis is drawn as a gap.

Capability evidence

Agentic78.5

Coding77.6

Knowledge70.6

Reasoning82.5

Multimodal & Grounded74.6

Instruction FollowingUnavailable

Runtime service evidence

Measured values with the date they were observed. Nothing is inferred from a sibling model or a provider claim.

Time to first token: latency to the first answer chunk (Artificial Analysis, via BenchLM). Reasoning models include thinking time, so values can run to tens or hundreds of seconds.

Evidence key: Observed

Runtime measurements
Measurement	Value	Observed	Last good	Evidence
Time to first token	24.09 s	2026-10-23	2026-10-23
Throughput	93 tok/s	2026-10-23	2026-10-23

Regional or per-endpoint measurements appear only when the API supplies them; none are modelled here.

Endpoint and price matrix

Every published price component, including cache reads and writes.

Price components
Component	USD	Evidence
Input / 1M tokens	$1.00
Output / 1M tokens	$2.70
Cache read / 1M tokens	$0.05
Cache write / 1M tokens	Unavailable
Blended / 1M (75% input / 25% output)	$1.43
Cost per successful task (LiveBench)	Unavailable

Self-hosted listingStepFun lists `step-5-preview` at $1.00 per million cache-miss input tokens, $0.05 per million cache-hit input tokens, and $2.70 per million output tokens. Output tokens include both the reasoning process and final answer. StepFun now publishes self-hostable FP8 and BF16 weights under StepFun Model License 1.0. These prices describe the hosted API; running the weights requires hardware.The rates above are a hosted price matched from another provider, not a first-party list price.

Workload-aware monthly cost example

10 conversations per day × 8 messages × 22 active days, 1200 input and 400 output tokens per message, no cache. This uses the same calculator as the cost simulator, so an unavailable applicable rate makes the example unavailable too.

Modelled monthly cost$4.01

Modelled tokens2.82M

Open the simulatorChange this workload

Benchmark record

43 matched benchmark rows with their published value, unit, and provenance.

Knowledge

Knowledge benchmarks
Benchmark	Value	Tasks	Difficulty	Provenance
GPQA-D GPQA Diamond	93.5	Graduate-level science questions	Graduate level	GPQA: A Graduate-Level Google-Proof Q&A Benchmark
HLE Humanity's Last Exam	46.5	Expert-level questions	Frontier expert level	Humanity's Last Exam
Artificial Analysis Intelligence Index Artificial Analysis Intelligence Index	43.7	Cross-benchmark intelligence index	Display-only external reference	Artificial Analysis
AA-HLE Artificial Analysis Humanity's Last Exam	46.5

Lifecycle and limitations log

Lifecycle events the source associates with this model, plus what this profile does not claim.

Step 5 PreviewRelease · confirmed · announced Unavailable · effective 2026-09-20StepFun launch announcement

What this profile does not claimValues are reproduced exactly as their sources published them, in the units those sources declared; none are converted, interpolated, or averaged across providers. Any field marked unavailable was attempted and not returned — the attempt timestamp is in each badge. Last attempted fetch for this model’s score: 2026-10-23 13:17 UTC.

Benchmark

Value

Tasks

Difficulty

Provenance

DRACO

Data Research and Analysis with Complex Operations

83.3

Agentic data research and analysis tasks

Professional data analysis

Claude Opus 5 System Card

Terminal-Bench 4.0

Terminal-Bench 4.0

33.30

66 professional computer-work tasks

Frontier autonomous knowledge work

Terminal-Bench 4.0

AA Briefcase

Artificial Analysis Briefcase

1424

Professional knowledge-work tasks

Professional work

Artificial Analysis Briefcase Benchmark Leaderboard

AA AutomationBench

Artificial Analysis AutomationBench

51.0

Business-process automation tasks

Agentic automation

Artificial Analysis AutomationBench Benchmark Leaderboard

AA EnterpriseOps-Gym

Artificial Analysis EnterpriseOps-Gym

47.2

Enterprise operations workflows

Enterprise agent operations

Artificial Analysis EnterpriseOps-Gym Benchmark Leaderboard

AA Harvey LAB v1.0

Artificial Analysis Harvey LAB-AA v1.0

93.4

Legal agent tasks

Professional legal work

Artificial Analysis Harvey LAB-AA Benchmark Leaderboard

AA ITBench

Artificial Analysis ITBench-AA

55.6

IT incident-response tasks

Enterprise IT operations

Artificial Analysis ITBench-AA Benchmark Leaderboard

Terminal-Bench 2.1

Terminal-Bench 2.1

85.0

Terminal-based software-agent tasks

Professional software engineering

Terminal-Bench

BrowseComp

BrowseComp

88.7

Research questions requiring browsing

Hard web research

BrowseComp

GDPval-AA

GDPval-AA

1566

Agentic real-world work tasks

Professional agentic workflows

DeepSeek-V4 Technical Report

GDPval-AA

GDPval-AA normalized

54.3

Economically valuable tasks

Professional agentic workflows

Artificial Analysis model benchmarks

AA Terminal-Bench 4.0

Artificial Analysis Terminal-Bench v4.0

33.3

Terminal-based agent tasks

Agentic software engineering

Artificial Analysis Terminal-Bench v4.0 Benchmark Leaderboard

GDP.pdf

Artificial Analysis GDP.pdf

14.8

Professional document-production tasks

Professional knowledge work

GDP.pdf Benchmark Leaderboard

AA-AnalystAgent

Artificial Analysis AnalystAgent

35.0

Spreadsheet and document analysis questions

Business and data analysis

AA-AnalystAgent Benchmark Leaderboard

APEX-Agents-AA

APEX-Agents-AA

38.0

452 professional-services agent tasks

Long-horizon workplace agent tasks

APEX-Agents-AA Benchmark Leaderboard

CyberGym

CyberGym

84.7

1,507 vulnerability analysis instances

Real-world cybersecurity

CyberGym: Evaluating AI Agents' Real-World Cybersecurity Capabilities at Scale

JobBench

JobBench

59.0

130 tasks across 35 occupations

Professional multi-source workflows

JobBench: Aligning Agent Work With Human Will

MCP Atlas

MCP Atlas

85.6

Tool-integrated agent tasks

Advanced tool use

Introducing GPT-5.4 mini and nano

Toolathlon-Verified

Toolathlon-Verified

74.1

Verified multi-tool workflows

Advanced tool use

Kimi K3: Open Frontier Intelligence

AutomationBench

AutomationBench

44.0

600 public automation tasks

Long-horizon automation

Kimi K3: Open Frontier Intelligence

Agents' Last Exam

Agents' Last Exam

29.5

Agent tasks

Advanced agentic work

DeepSeek V4 Flash 0731 update

APEX-Agents

APEX-Agents

37.8

Professional-services agent tasks

Long-horizon professional work

Kimi K3: Open Frontier Intelligence

Hermes Index

Hermes Index: agentic capability and cost per task inside Hermes Agent

32.8

Four suites: Hermes Bench (150 tasks), TerminalBench 4 without its 4 GPU tasks, TerminalBench Science and SkillsBench

Frontier agentic

Hermes Index: an agentic model leaderboard

Hermes Bench

Hermes Bench: 150 agent tasks graded on the files and state the agent leaves behind

69.2

150 tasks across 25 categories

Frontier agentic

Hermes Bench (Hermes Index)

Benchmark	Value	Tasks	Difficulty	Provenance
Terminal-Bench 2.1 Terminal-Bench 2.1	85.0	Terminal-based software-agent tasks	Professional software engineering	Terminal-Bench
ProgramBench ProgramBench: Can Language Models Rebuild Programs From Scratch?	80.5	200 program reconstruction tasks	Full-repository software architecture	ProgramBench: Can Language Models Rebuild Programs From Scratch?
MLS-Bench Lite MLS-Bench Lite	40.5	30 machine-learning research tasks	ML research and systems engineering	MLS-Bench
SciCode Scientific Code Benchmark	58.9	—	—	BenchLM
AA-SciCode Artificial Analysis SciCode	58.9	Scientific coding subproblems	Scientific programming	Artificial Analysis SciCode Benchmark Leaderboard
DeepSWE DeepSWE	67.7	113 software engineering tasks across 91 repositories and 5 languages	Long-horizon software engineering	DeepSWE benchmark blog

Benchmark	Value	Tasks	Difficulty	Provenance
AA-LCR Artificial Analysis Long Context Reasoning	88.3	Long-context reasoning tasks	Long-context reasoning	Artificial Analysis model benchmarks
MLCR-AA Medical Long Context Reasoning (MLCR-AA)	16.7	Long, fragmented medical-record reasoning	Long-context medical reasoning	Medical Long Context Reasoning (MLCR-AA)
CritPt Critical Physics Tasks	20.9	Research-level physics questions	Research-level physics reasoning	CritPt Benchmark Leaderboard

Benchmark	Value	Tasks	Difficulty	Provenance
MMMU-Pro Massive Multi-discipline Multimodal Understanding Pro	76	Multimodal academic reasoning	Frontier multimodal	MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark
AA-MMMU-Pro Artificial Analysis MMMU-Pro	76.4	Multimodal academic reasoning	Frontier multimodal	Artificial Analysis MMMU-Pro Benchmark Leaderboard
OfficeQA Pro OfficeQA Pro	60.3	Document and spreadsheet tasks	Enterprise grounded reasoning	OfficeQA Pro: An Enterprise Benchmark for End-to-End Grounded Reasoning

Explore models

Top 1-5

Top 6-10

Featured

Capabilities

Benchmarks

Compare

Featured model comparisons

Featured provider comparisons

Articles

Explore models

Top 1-5

Top 6-10

Featured

Capabilities

Benchmarks

Compare

Featured model comparisons

Featured provider comparisons

Articles

Step 5 Preview

Capability shape

Capability evidence

Runtime service evidence

Endpoint and price matrix

Workload-aware monthly cost example

Benchmark record

Knowledge

Lifecycle and limitations log

Coding

Reasoning

Agentic

Multimodal & Grounded

Benchmark	Value	Tasks	Difficulty	Provenance
DRACO Data Research and Analysis with Complex Operations	83.3	Agentic data research and analysis tasks	Professional data analysis	Claude Opus 5 System Card
Terminal-Bench 4.0 Terminal-Bench 4.0	33.30	66 professional computer-work tasks	Frontier autonomous knowledge work	Terminal-Bench 4.0
AA Briefcase Artificial Analysis Briefcase	1424	Professional knowledge-work tasks	Professional work	Artificial Analysis Briefcase Benchmark Leaderboard
AA AutomationBench Artificial Analysis AutomationBench	51.0	Business-process automation tasks	Agentic automation	Artificial Analysis AutomationBench Benchmark Leaderboard
AA EnterpriseOps-Gym Artificial Analysis EnterpriseOps-Gym	47.2	Enterprise operations workflows	Enterprise agent operations	Artificial Analysis EnterpriseOps-Gym Benchmark Leaderboard
AA Harvey LAB v1.0 Artificial Analysis Harvey LAB-AA v1.0	93.4	Legal agent tasks	Professional legal work	Artificial Analysis Harvey LAB-AA Benchmark Leaderboard
AA ITBench Artificial Analysis ITBench-AA	55.6	IT incident-response tasks	Enterprise IT operations	Artificial Analysis ITBench-AA Benchmark Leaderboard
Terminal-Bench 2.1 Terminal-Bench 2.1	85.0	Terminal-based software-agent tasks	Professional software engineering	Terminal-Bench
BrowseComp BrowseComp	88.7	Research questions requiring browsing	Hard web research	BrowseComp
GDPval-AA GDPval-AA	1566	Agentic real-world work tasks	Professional agentic workflows	DeepSeek-V4 Technical Report
GDPval-AA GDPval-AA normalized	54.3	Economically valuable tasks	Professional agentic workflows	Artificial Analysis model benchmarks
AA Terminal-Bench 4.0 Artificial Analysis Terminal-Bench v4.0	33.3	Terminal-based agent tasks	Agentic software engineering	Artificial Analysis Terminal-Bench v4.0 Benchmark Leaderboard
GDP.pdf Artificial Analysis GDP.pdf	14.8	Professional document-production tasks	Professional knowledge work	GDP.pdf Benchmark Leaderboard
AA-AnalystAgent Artificial Analysis AnalystAgent	35.0	Spreadsheet and document analysis questions	Business and data analysis	AA-AnalystAgent Benchmark Leaderboard
APEX-Agents-AA APEX-Agents-AA	38.0	452 professional-services agent tasks	Long-horizon workplace agent tasks	APEX-Agents-AA Benchmark Leaderboard
CyberGym CyberGym	84.7	1,507 vulnerability analysis instances	Real-world cybersecurity	CyberGym: Evaluating AI Agents' Real-World Cybersecurity Capabilities at Scale
JobBench JobBench	59.0	130 tasks across 35 occupations	Professional multi-source workflows	JobBench: Aligning Agent Work With Human Will
MCP Atlas MCP Atlas	85.6	Tool-integrated agent tasks	Advanced tool use	Introducing GPT-5.4 mini and nano
Toolathlon-Verified Toolathlon-Verified	74.1	Verified multi-tool workflows	Advanced tool use	Kimi K3: Open Frontier Intelligence
AutomationBench AutomationBench	44.0	600 public automation tasks	Long-horizon automation	Kimi K3: Open Frontier Intelligence
Agents' Last Exam Agents' Last Exam	29.5	Agent tasks	Advanced agentic work	DeepSeek V4 Flash 0731 update
APEX-Agents APEX-Agents	37.8	Professional-services agent tasks	Long-horizon professional work	Kimi K3: Open Frontier Intelligence
Hermes Index Hermes Index: agentic capability and cost per task inside Hermes Agent	32.8	Four suites: Hermes Bench (150 tasks), TerminalBench 4 without its 4 GPU tasks, TerminalBench Science and SkillsBench	Frontier agentic	Hermes Index: an agentic model leaderboard
Hermes Bench Hermes Bench: 150 agent tasks graded on the files and state the agent leaves behind	69.2	150 tasks across 25 categories	Frontier agentic	Hermes Bench (Hermes Index)