What are Step 3's benchmark scores?

As of 24 Oct 2026, Step 3 scores 51.5 ±2.1 on the BenchLeader Index, strongest in reasoning (55) and weakest in multimodal (49). That rests on 6 independent benchmark results, and every score on this page links to its publisher.

How does Step 3 rank?

Step 3 is #345 of 785 ranked configurations on 24 Oct 2026. 291 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.

StepFun Compare with another model

Step 3

Step 3 is a StepFun proprietary model. Its best configuration ranks #345 of 785 on the BenchLeader Index at 51.5 ±2.1, in the upper half. The ± is the point: 291 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in reasoning (55) and lowest in multimodal (49). Last measured 22 Oct 2026.

Blended price: –
Output speed: –
First answer: –
Context: –
Full answer: –
Cost per run: –
Released: –; date not published by our sources

How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.

Overall index52
Reasoning55
Coding53
Maths53
Human preference51
Multimodal49

Versions

StepFun has shipped 2 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.

Model	Released	Index	Rank
Step 2	1 Jan 2025	–	–
Step 3this page	–	51.5	#345

How its index has moved

9 Sept 2026 to 21 Oct 2026

The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.

Benchmark results

See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Reasoning

Benchmark	not stated	Source	Trend
Arena Hard Prompts	1379#188	Arena (LMArena)
Kagi LLM Benchmark	62.3%#45	Kagi LLM Benchmark

Coding

Benchmark	not stated	Source	Trend
Arena Coding	1409#187	Arena (LMArena)

Maths

Benchmark	not stated	Source	Trend
Arena Maths	1362#189	Arena (LMArena)

Instruction following

Benchmark	not stated	Source	Trend
Arena Instruction Followingnot in index	1345#193	Arena (LMArena)

Human preference

Benchmark	not stated	Source
Arena Text	1349#202	Arena (LMArena)
Arena Creative Writingnot in index	1310#204	Arena (LMArena)
Arena Multi-turnnot in index	1344#202	Arena (LMArena)
Arena Longer Queriesnot in index	1345#207	Arena (LMArena)

Multimodal

Benchmark	not stated	Source	Trend
Arena Vision	1176#103	Arena (LMArena)

New this month

Questions people ask

What are Step 3's benchmark scores?: As of 24 Oct 2026, Step 3 scores 51.5 ±2.1 on the BenchLeader Index, strongest in reasoning (55) and weakest in multimodal (49). That rests on 6 independent benchmark results, and every score on this page links to its publisher.
How does Step 3 rank?: Step 3 is #345 of 785 ranked configurations on 24 Oct 2026. 291 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.

Data as of 24 Oct 2026. Compare with another model.

Cite as: BenchLeader, “Step 3: benchmarks, pricing, speed and rank”, data as of 24 Oct 2026, https://www.benchleader.com/models/step-3.