Model comparison

GPT-5.2 Pro vs Step 3.5 Flash

GPT-5.2 Pro is the stronger model overall, scoring 52.3 to 42.3 on the Noometry Index. Step 3.5 Flash costs 385× less per token, which makes it the better buy when GPT-5.2 Pro's lead doesn't matter for your workload.

Last verified October 11, 2026. 1 shared benchmarks.

GPT-5.2 Pro OpenAI

52.3

Rank #39 Reported

Step 3.5 Flash

StepFun

42.3

Rank #116 Confirmed

Summary

They share 1 benchmark with published results for both. GPT-5.2 Pro scores higher in 2 categories and Step 3.5 Flash in 0 categories; 2 gaps are clear of the uncertainty.
The widest gap is in reasoning, where GPT-5.2 Pro leads 51.5 to 22.2.
The biggest single-benchmark swing is NYT Connections (extended): 79.3% for GPT-5.2 Pro and 28.4% for Step 3.5 Flash.
Step 3.5 Flash is cheaper at $0.10 / $0.30 per million input/output tokens, against $21 / $168 for GPT-5.2 Pro.
GPT-5.2 Pro accepts more context: 400K tokens versus 256K.
Step 3.5 Flash has downloadable open weights; the other is API-only.

Side by side

GPT-5.2 Pro and Step 3.5 Flash specifications
	GPT-5.2 Pro	Step 3.5 Flash
Provider	OpenAI	StepFun
Noometry Index	52.3	42.3
Released	2025-12-11	2026-01-29
Weights	Proprietary	Open
Context window	400K	256K
Max output	128K	256K
Input $ / M tokens	$21	$0.10
Output $ / M tokens	$168	$0.30
Results tracked	8	19

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding Not comparable

GPT-5.2 Pro: —, Step 3.5 Flash: 42.4 (#105)

Coding benchmarks
Benchmark	GPT-5.2 Pro	Step 3.5 Flash
LMArena Coding	—	1436

Reasoning GPT-5.2 Pro leads

GPT-5.2 Pro: 51.5 (#33), Step 3.5 Flash: 22.2 (#202)

Reasoning benchmarks
Benchmark	GPT-5.2 Pro	Step 3.5 Flash
NYT Connections (extended)	79.3%	28.4%
ARC-AGI-2	54.2%	—
SimpleBench	57.4%	—
ARC-AGI-1	90.5%	—
LMArena Hard Prompts	—	1411
Epoch Capabilities Index	155.4	—

Math GPT-5.2 Pro leads

GPT-5.2 Pro: 65.3 (#29), Step 3.5 Flash: 42.6 (#84)

Math benchmarks
Benchmark	GPT-5.2 Pro	Step 3.5 Flash
FrontierMath (Tiers 1-3)	74%	—
FrontierMath Tier 4	46%	—
MathArena Final-Answer Competitions	—	66.8%
LMArena Math	—	1408
FrontierMath Tier 4 (v1)	31.3%	—

Knowledge Not comparable

GPT-5.2 Pro: —, Step 3.5 Flash: 39.6 (#132)

Knowledge benchmarks
Benchmark	GPT-5.2 Pro	Step 3.5 Flash
LMArena Expert	—	1421

Multilingual Not comparable

GPT-5.2 Pro: —, Step 3.5 Flash: 50.5 (#119)

Multilingual benchmarks
Benchmark	GPT-5.2 Pro	Step 3.5 Flash
LMArena Non-English	—	1385
LMArena Chinese	—	1447
LMArena French	—	1421
LMArena German	—	1405
LMArena Japanese	—	1354
LMArena Korean	—	1352
LMArena Russian	—	1385
LMArena Spanish	—	1419

Instruction Following Not comparable

GPT-5.2 Pro: —, Step 3.5 Flash: 73.1 (#124)

Instruction Following benchmarks
Benchmark	GPT-5.2 Pro	Step 3.5 Flash
LMArena Instruction Following	—	1385

Long Context Not comparable

GPT-5.2 Pro: —, Step 3.5 Flash: 42.8 (#117)

Long Context benchmarks
Benchmark	GPT-5.2 Pro	Step 3.5 Flash
LMArena Longer Query	—	1402

Writing & Preference Not comparable

GPT-5.2 Pro: —, Step 3.5 Flash: 58.8 (#113)

Writing & Preference benchmarks
Benchmark	GPT-5.2 Pro	Step 3.5 Flash
LMArena Text	—	1403
LMArena Creative Writing	—	1357
LMArena Multi-Turn	—	1405

Frequently asked questions

Is GPT-5.2 Pro better than Step 3.5 Flash?

Which is cheaper, GPT-5.2 Pro or Step 3.5 Flash?

Step 3.5 Flash is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; GPT-5.2 Pro lists at $21 and $168.

Which has the bigger context window?

GPT-5.2 Pro does, with 400K tokens against 256K.

How many benchmarks do GPT-5.2 Pro and Step 3.5 Flash share?

1 benchmark has published results for both models. GPT-5.2 Pro has 8 scored results on Noometry and Step 3.5 Flash has 19.