Noometry

Model comparison

Qwen3.8 27B vs Step 5 Preview

Step 5 Preview is the stronger model overall, scoring 47.9 to 46.0 on the Noometry Index.

Last verified . 18 shared benchmarks.

Qwen3.8 27B Alibaba (Qwen)

46.0

Rank #68 Confirmed

Step 5 Preview StepFun

47.9

Rank #58 Confirmed

Summary

  • They share 18 benchmarks with published results for both. Qwen3.8 27B scores higher in 4 categories and Step 5 Preview in 5 categories; 3 gaps are clear of the uncertainty.
  • The widest gap is in math, where Step 5 Preview leads 46.1 to 37.1.
  • The biggest single-benchmark swing is ProofBench: 16% for Qwen3.8 27B and 42% for Step 5 Preview.
  • Qwen3.8 27B is cheaper at $0.99 / $1.49 per million input/output tokens, against $1 / $2.70 for Step 5 Preview.
  • Step 5 Preview accepts more context: 1.02M tokens versus 262K.
  • Qwen3.8 27B has downloadable open weights; the other is API-only.

Side by side

Qwen3.8 27BStep 5 Preview
ProviderAlibaba (Qwen)StepFun
Noometry Index46.047.9
Released2026-08-142026-09-16
WeightsOpenProprietary
Context window262K1.02M
Max output33K66K
Input $ / M tokens$0.99$1
Output $ / M tokens$1.49$2.70
Results tracked3118

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

Category by category

Coding Step 5 Preview leads

Qwen3.8 27B: 50.5 (#44), Step 5 Preview: 51.8 (#36)

BenchmarkQwen3.8 27BStep 5 Preview
LMArena WebDev15931564
SciCode46.6%58.9%
LMArena Coding14821480

Agentic & Tool Use Not comparable

Qwen3.8 27B: 32.9 (#57), Step 5 Preview: —

BenchmarkQwen3.8 27BStep 5 Preview
APEX-Agents47.5%—

Reasoning Too close to call

Qwen3.8 27B: 41.0 (#54), Step 5 Preview: 40.0 (#57)

BenchmarkQwen3.8 27BStep 5 Preview
CritPt5.4%20.9%
LMArena Hard Prompts14601465
ARC-AGI-242.4%—
NYT Connections (extended)54.5%—
ARC-AGI-187.5%—
DTBench88%—
LMCA41.4%—
Surface Evolver Bench45%—
Epoch Capabilities Index149.38—

Math Step 5 Preview leads

Qwen3.8 27B: 37.1 (#161), Step 5 Preview: 46.1 (#72)

BenchmarkQwen3.8 27BStep 5 Preview
ProofBench16%42%
LMArena Math14561470

Knowledge Too close to call

Qwen3.8 27B: 41.6 (#109), Step 5 Preview: 41.2 (#112)

BenchmarkQwen3.8 27BStep 5 Preview
LMArena Expert14821470

Multimodal Too close to call

Qwen3.8 27B: 41.3 (#37), Step 5 Preview: 41.0 (#41)

BenchmarkQwen3.8 27BStep 5 Preview
LMArena Vision12711267

Multilingual Too close to call

Qwen3.8 27B: 53.7 (#60), Step 5 Preview: 53.8 (#54)

BenchmarkQwen3.8 27BStep 5 Preview
LMArena Non-English14301432
LMArena Chinese15041519
LMArena Russian14151436
LMArena Spanish14481439
LMArena French1465—
LMArena German1438—
LMArena Japanese1384—
LMArena Korean1393—

Instruction Following Too close to call

Qwen3.8 27B: 75.8 (#53), Step 5 Preview: 76.0 (#50)

BenchmarkQwen3.8 27BStep 5 Preview
LMArena Instruction Following14391444

Long Context Too close to call

Qwen3.8 27B: 44.3 (#70), Step 5 Preview: 44.5 (#64)

BenchmarkQwen3.8 27BStep 5 Preview
LMArena Longer Query14501455

Writing & Preference Qwen3.8 27B leads

Qwen3.8 27B: 65.8 (#43), Step 5 Preview: 62.8 (#71)

BenchmarkQwen3.8 27BStep 5 Preview
LMArena Text14411442
LMArena Creative Writing13841410
LMArena Multi-Turn14411447
EQ-Bench Creative Writing1671—

Frequently asked questions

Is Qwen3.8 27B better than Step 5 Preview?

Step 5 Preview is the stronger model overall, scoring 47.9 to 46.0 on the Noometry Index.

Which is cheaper, Qwen3.8 27B or Step 5 Preview?

Qwen3.8 27B is cheaper. It lists at $0.99 per million input tokens and $1.49 per million output tokens; Step 5 Preview lists at $1 and $2.70.

Is Qwen3.8 27B or Step 5 Preview better for coding?

Step 5 Preview scores higher on coding benchmarks: 51.8 versus 50.5 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 Qwen3.8 27B and Step 5 Preview share?

18 benchmarks have published results for both models. Qwen3.8 27B has 31 scored results on Noometry and Step 5 Preview has 18.

Related comparisons

Go deeper