Skip to content
aiqbench

Stepfun · model passport

StepFun: Step 3.5 Flash

step-3-5-flash
  • Open weights
  • Apache 2.0 licence
  • 199.4B params
  • 262,144 context
  • Released 2026-02-01(HF repo created)
6/27axes covered
Overall · LMArena
1,393#158 of 396
Best percentile
73.5IFBench (Artificial Analysis)
Cohort standing
54.9median 56.4

What the data says

Benchmarked on 6 of 27 tracked axes. Every figure below is the model's own published result on that board.

Uneven: 47.9 percentile points between its best and worst axis — 73.5 pct on IFBench (Artificial Analysis) versus 25.6 pct on Factuality Arena.

  • IFBench (Artificial Analysis)73.5 pct64.6%Verify →
  • Coding63.4 pct1,449Verify →
  • Instruction following61.4 pct1,384Verify →
  • Overall60.4 pct1,393Verify →
  • Hard prompts59.6 pct1,412Verify →
  • Factuality Arena25.6 pct1,416Verify →
  • WebDev ArenaNot benchmarked
  • Vision ArenaNot benchmarked
  • Search ArenaNot benchmarked
  • Document ArenaNot benchmarked
  • Agent 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
  • 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-3.5-Flash
Parameters199.4B
ArchitectureStep3p5ForCausalLM
LicenceApache 2.0

Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "stepfun-ai/Step-3.5-Flash"
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-3.5-flash
Context length262,144 tokens
Max output65,536 tokens
Input price$0.10 per 1M tokens
Output price$0.30 per 1M tokens

API parameters this endpoint supports

Frequency penaltyInclude reasoningMax tokensReasoningTemperatureTool choiceTool callingTop-kTop-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-3.5-flash","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-3.5-flash stands at 54.9 (raw mean 57.3) — measured on 6 of 27 axes.

    standingtemplateVerify →

  • step-3.5-flash stands at 54.9 versus a cohort median of 56.4 across 110 models released within 180 days — below 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 →