Noometry

Model comparison

Llama 3.1-405B vs Step 5 Preview

Step 5 Preview is the stronger model overall, scoring 47.9 to 30.7 on the Noometry Index.

Last verified . 13 shared benchmarks.

Llama 3.1-405B Meta

30.7

Rank #288 Confirmed

Step 5 Preview StepFun

47.9

Rank #58 Confirmed

Summary

  • They share 13 benchmarks with published results for both. Llama 3.1-405B scores higher in 0 categories and Step 5 Preview in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where Step 5 Preview leads 46.1 to 18.4.
  • Llama 3.1-405B has downloadable open weights; the other is API-only.

Side by side

Llama 3.1-405BStep 5 Preview
ProviderMetaStepFun
Noometry Index30.747.9
Released2024-07-232026-09-16
WeightsOpenProprietary
Context window—1.02M
Max output—66K
Input $ / M tokens—$1
Output $ / M tokens—$2.70
Results tracked4218

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

Category by category

Coding Step 5 Preview leads

Llama 3.1-405B: 33.1 (#262), Step 5 Preview: 51.8 (#36)

BenchmarkLlama 3.1-405BStep 5 Preview
LMArena Coding12911480
LMArena WebDev—1564
SciCode—58.9%
WeirdML21.4%—

Agentic & Tool Use Not comparable

Llama 3.1-405B: 21.0 (#140), Step 5 Preview: —

BenchmarkLlama 3.1-405BStep 5 Preview
TheAgentCompany7.4%—
Cybench7.5%—

Reasoning Step 5 Preview leads

Llama 3.1-405B: 16.8 (#300), Step 5 Preview: 40.0 (#57)

BenchmarkLlama 3.1-405BStep 5 Preview
LMArena Hard Prompts12691465
SimpleBench23%—
Kagi LLM Benchmark45%—
CritPt—20.9%
DTBench61.4%—
BIG-Bench Hard82.9%—
Epoch Capabilities Index128.75—
ForecastBench59.9—
HellaSwag89.2%—
PIQA85.9%—
WinoGrande89.2%—

Math Step 5 Preview leads

Llama 3.1-405B: 18.4 (#290), Step 5 Preview: 46.1 (#72)

BenchmarkLlama 3.1-405BStep 5 Preview
LMArena Math12811470
OTIS Mock AIME 2024-20259.7%—
ProofBench—42%
Omni-MATH24.9%—
MATH Level 549.8%—

Knowledge Step 5 Preview leads

Llama 3.1-405B: 30.4 (#227), Step 5 Preview: 41.2 (#112)

BenchmarkLlama 3.1-405BStep 5 Preview
LMArena Expert12431470
GPQA Diamond50.9%—
MMLU-Pro72.3%—
Confabulations17.6%—
GPQA (HELM)52.2%—
ARC (AI2) Challenge95.3%—
MMLU84.5%—
TriviaQA82.7%—

Multimodal Not comparable

Llama 3.1-405B: —, Step 5 Preview: 41.0 (#41)

BenchmarkLlama 3.1-405BStep 5 Preview
LMArena Vision—1267

Multilingual Step 5 Preview leads

Llama 3.1-405B: 40.7 (#214), Step 5 Preview: 53.8 (#54)

BenchmarkLlama 3.1-405BStep 5 Preview
LMArena Non-English12481432
LMArena Chinese12421519
LMArena Russian12651436
LMArena Spanish12601439
LMArena French1279—
LMArena German1252—
LMArena Japanese1208—
LMArena Korean1184—

Instruction Following Step 5 Preview leads

Llama 3.1-405B: 65.9 (#214), Step 5 Preview: 76.0 (#50)

BenchmarkLlama 3.1-405BStep 5 Preview
LMArena Instruction Following12591444
IFEval81.1%—

Long Context Step 5 Preview leads

Llama 3.1-405B: 38.4 (#197), Step 5 Preview: 44.5 (#64)

BenchmarkLlama 3.1-405BStep 5 Preview
LMArena Longer Query12661455

Writing & Preference Step 5 Preview leads

Llama 3.1-405B: 38.9 (#251), Step 5 Preview: 62.8 (#71)

BenchmarkLlama 3.1-405BStep 5 Preview
LMArena Text12841442
LMArena Creative Writing12621410
LMArena Multi-Turn12971447
EQ-Bench Creative Writing870—
WildBench78.3%—

Frequently asked questions

Is Llama 3.1-405B better than Step 5 Preview?

Step 5 Preview is the stronger model overall, scoring 47.9 to 30.7 on the Noometry Index.

Is Llama 3.1-405B or Step 5 Preview better for coding?

Step 5 Preview scores higher on coding benchmarks: 51.8 versus 33.1 in the Noometry coding category.

How many benchmarks do Llama 3.1-405B and Step 5 Preview share?

13 benchmarks have published results for both models. Llama 3.1-405B has 42 scored results on Noometry and Step 5 Preview has 18.

Related comparisons

Go deeper