Noometry

Model comparison

Qwen3 Max vs Step 5 Preview

Step 5 Preview is the stronger model overall, scoring 47.9 to 43.7 on the Noometry Index.

Last verified . 13 shared benchmarks.

Qwen3 Max Alibaba (Qwen)

43.7

Rank #87 Confirmed

Step 5 Preview StepFun

47.9

Rank #58 Confirmed

Summary

  • They share 13 benchmarks with published results for both. Qwen3 Max scores higher in 1 category and Step 5 Preview in 7 categories; 6 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Step 5 Preview leads 40.0 to 22.6.
  • Step 5 Preview is cheaper at $1 / $2.70 per million input/output tokens, against $1.20 / $6 for Qwen3 Max.
  • Step 5 Preview accepts more context: 1.02M tokens versus 262K.

Side by side

Qwen3 MaxStep 5 Preview
ProviderAlibaba (Qwen)StepFun
Noometry Index43.747.9
Released2025-09-232026-09-16
WeightsProprietaryProprietary
Context window262K1.02M
Max output66K66K
Input $ / M tokens$1.20$1
Output $ / M tokens$6$2.70
Results tracked3318

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

Category by category

Coding Step 5 Preview leads

Qwen3 Max: 43.0 (#93), Step 5 Preview: 51.8 (#36)

BenchmarkQwen3 MaxStep 5 Preview
LMArena Coding14561480
LMArena WebDev—1564
SciCode—58.9%
ALE-Bench370.45—

Agentic & Tool Use Not comparable

Qwen3 Max: —, Step 5 Preview: —

BenchmarkQwen3 MaxStep 5 Preview
Vending-Bench 271.56—

Reasoning Step 5 Preview leads

Qwen3 Max: 22.6 (#190), Step 5 Preview: 40.0 (#57)

BenchmarkQwen3 MaxStep 5 Preview
LMArena Hard Prompts14481465
Kagi LLM Benchmark72.5%—
NYT Connections (extended)30.1%—
CritPt—20.9%
Chess Puzzles4%—
Mystery Game Puzzles5%—
DTBench82.1%—
LMCA28.3%—
Epoch Capabilities Index142.38—

Math Step 5 Preview leads

Qwen3 Max: 38.7 (#131), Step 5 Preview: 46.1 (#72)

BenchmarkQwen3 MaxStep 5 Preview
LMArena Math14461470
FrontierMath (Tiers 1-3)18.9%—
OTIS Mock AIME 2024-202573.3%—
ProofBench—42%
MATH Level 597.1%—

Knowledge Qwen3 Max leads

Qwen3 Max: 48.1 (#78), Step 5 Preview: 41.2 (#112)

BenchmarkQwen3 MaxStep 5 Preview
LMArena Expert14551470
GPQA Diamond72.6%—
SimpleQA Verified48.7%—

Multimodal Not comparable

Qwen3 Max: —, Step 5 Preview: 41.0 (#41)

BenchmarkQwen3 MaxStep 5 Preview
LMArena Vision—1267

Multilingual Too close to call

Qwen3 Max: 53.7 (#62), Step 5 Preview: 53.8 (#54)

BenchmarkQwen3 MaxStep 5 Preview
LMArena Non-English14291432
LMArena Chinese14781519
LMArena Russian14281436
LMArena Spanish14621439
LMArena French1449—
LMArena German1463—
LMArena Japanese1397—
LMArena Korean1399—

Instruction Following Step 5 Preview leads

Qwen3 Max: 74.8 (#87), Step 5 Preview: 76.0 (#50)

BenchmarkQwen3 MaxStep 5 Preview
LMArena Instruction Following14191444

Long Context Step 5 Preview leads

Qwen3 Max: 41.6 (#134), Step 5 Preview: 44.5 (#64)

BenchmarkQwen3 MaxStep 5 Preview
LMArena Longer Query14381455
Fiction.LiveBench66.7%—
CL-bench14.5%—

Writing & Preference Too close to call

Qwen3 Max: 62.4 (#76), Step 5 Preview: 62.8 (#71)

BenchmarkQwen3 MaxStep 5 Preview
LMArena Text14391442
LMArena Creative Writing14021410
LMArena Multi-Turn14461447

Frequently asked questions

Is Qwen3 Max better than Step 5 Preview?

Step 5 Preview is the stronger model overall, scoring 47.9 to 43.7 on the Noometry Index.

Which is cheaper, Qwen3 Max or Step 5 Preview?

Step 5 Preview is cheaper. It lists at $1 per million input tokens and $2.70 per million output tokens; Qwen3 Max lists at $1.20 and $6.

Is Qwen3 Max or Step 5 Preview better for coding?

Step 5 Preview scores higher on coding benchmarks: 51.8 versus 43.0 in the Noometry coding category.

Which has the bigger context window?

Step 5 Preview does, with 1.02M tokens against 262K.

How many benchmarks do Qwen3 Max and Step 5 Preview share?

13 benchmarks have published results for both models. Qwen3 Max has 33 scored results on Noometry and Step 5 Preview has 18.

Related comparisons

Go deeper