Skip to content
aiqbench

Stepfun · model passport

StepFun: Step 5 Preview

step-5-preview
  • Open weights
  • Proprietary licence
  • 604.3B params
  • 1,048,576 context
  • Released 2026-09-20(HF repo created)
8/27axes covered
Overall · LMArena
1,447#74 of 396
Best percentile
85.7Coding
Cohort standing
66.2median 60.3

What the data says

Benchmarked on 8 of 27 tracked axes. Every figure below is the model's own published result on that board.

Strongest axes

  • 85.7 pct — Coding (1,508)

Uneven: 47.5 percentile points between its best and worst axis — 85.7 pct on Coding versus 38.2 pct on Agent Arena.

  • Coding85.7 pct1,508Verify →
  • Hard prompts84.1 pct1,475Verify →
  • Instruction following82.6 pct1,441Verify →
  • Overall81.6 pct1,447Verify →
  • WebDev Arena77 pct1,565Verify →
  • Vision Arena69.3 pct1,250Verify →
  • Factuality Arena59.9 pct1,446Verify →
  • Agent Arena38.2 pct-0.014Verify →
  • Search ArenaNot benchmarked
  • Document ArenaNot benchmarked
  • GPQA DiamondNot benchmarked
  • AIME (math)Not benchmarked
  • FrontierMathNot benchmarked
  • SimpleQA VerifiedNot benchmarked
  • ARC-AGI-2Not benchmarked
  • Humanity's Last ExamNot benchmarked
  • SimpleBenchNot benchmarked
  • LiveBenchNot benchmarked
  • BFCL v4 (function calling)Not benchmarked
  • SWE-bench Verified (live board)Not benchmarked
  • Terminal-Bench 4.0Not benchmarked
  • MMLU-Pro (Artificial Analysis)Not benchmarked
  • LiveCodeBench (Artificial Analysis)Not benchmarked
  • IFBench (Artificial Analysis)Not benchmarked
  • HHEM Hallucination (Vectara)Not benchmarked
  • SWE-bench VerifiedNot benchmarked
  • Terminal-BenchNot benchmarked

Percentiles are the model's standing within each board's own population. Axes marked "not benchmarked" have no published result for this model — that is an absence of data, not a low score.

How to run it

Run it yourself

Repositorystepfun-ai/Step-5-Preview
Parameters604.3B
ArchitectureStep5ForConditionalGeneration
LicenceProprietary

Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "stepfun-ai/Step-5-Preview"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

Ollama

We don't sync Ollama tags, so we won't print a command we can't verify — a wrong one would pull the wrong model. Search the Ollama library →

Unsloth fine-tuning recipes are model-specific and we don't derive them here. Unsloth docs →

Use it hosted

API model IDstepfun/step-5-preview
Context length1,048,576 tokens
Max output64,000 tokens
Acceptstext, image, video
Input price$1 per 1M tokens
Cached input price$0.10 per 1M tokens
Output price$2.70 per 1M tokens

API parameters this endpoint supports

Frequency penaltyInclude reasoningLogit biasLogprobsMax tokensMin-pPresence penaltyReasoningReasoning effortRepetition penaltyResponse formatSeedStop sequencesStructured outputsTemperatureTool choiceTool callingTop-kTop logprobsTop-p

Example request

curl https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"stepfun/step-5-preview","messages":[{"role":"user","content":"Hello"}]}'

View on OpenRouter →

Identifiers, limits and prices are synced from the upstream source shown above. Any field we haven't synced is left out entirely rather than guessed.

  • Step 5 Preview stands at 66.2 (raw mean 72.3) — measured on 8 of 27 axes.

    standingtemplateVerify →

  • Step 5 Preview stands at 66.2 versus a cohort median of 60.3 across 176 models released within 180 days — above the peer median.

    cohorttemplateVerify →

Every claim is a computed fact from the current board data, phrased by a low-temperature model call and rejected by a validation gate if any number is not in the source row. How the leaderboards work →