Noometry

Model comparison

Grok 4.20 Multi-Agent vs Step 5 Preview

Step 5 Preview is the stronger model overall, scoring 47.9 to 46.2 on the Noometry Index.

Last verified . 14 shared benchmarks.

Grok 4.20 Multi-Agent xAI

46.2

Rank #65 Confirmed

Step 5 Preview StepFun

47.9

Rank #58 Confirmed

Summary

  • They share 14 benchmarks with published results for both. Grok 4.20 Multi-Agent scores higher in 3 categories and Step 5 Preview in 6 categories; 5 gaps are clear of the uncertainty.
  • The widest gap is in coding, where Step 5 Preview leads 51.8 to 43.0.
  • Step 5 Preview is cheaper at $1 / $2.70 per million input/output tokens, against $1.25 / $2.50 for Grok 4.20 Multi-Agent.
  • Step 5 Preview accepts more context: 1.02M tokens versus 1M.

Side by side

Grok 4.20 Multi-AgentStep 5 Preview
ProviderxAIStepFun
Noometry Index46.247.9
Released2026-03-092026-09-16
WeightsProprietaryProprietary
Context window1M1.02M
Max output30K66K
Input $ / M tokens$1.25$1
Output $ / M tokens$2.50$2.70
Results tracked2018

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

Category by category

Coding Step 5 Preview leads

Grok 4.20 Multi-Agent: 43.0 (#92), Step 5 Preview: 51.8 (#36)

BenchmarkGrok 4.20 Multi-AgentStep 5 Preview
LMArena Coding14571480
LMArena WebDev—1564
SciCode—58.9%

Agentic & Tool Use Not comparable

Grok 4.20 Multi-Agent: —, Step 5 Preview: —

BenchmarkGrok 4.20 Multi-AgentStep 5 Preview
LMArena Search1204—

Reasoning Grok 4.20 Multi-Agent leads

Grok 4.20 Multi-Agent: 43.9 (#48), Step 5 Preview: 40.0 (#57)

BenchmarkGrok 4.20 Multi-AgentStep 5 Preview
LMArena Hard Prompts14481465
NYT Connections (extended)89.6%—
CritPt—20.9%

Math Step 5 Preview leads

Grok 4.20 Multi-Agent: 39.4 (#104), Step 5 Preview: 46.1 (#72)

BenchmarkGrok 4.20 Multi-AgentStep 5 Preview
LMArena Math14421470
ProofBench—42%

Knowledge Too close to call

Grok 4.20 Multi-Agent: 40.4 (#119), Step 5 Preview: 41.2 (#112)

BenchmarkGrok 4.20 Multi-AgentStep 5 Preview
LMArena Expert14451470

Multimodal Too close to call

Grok 4.20 Multi-Agent: 40.5 (#48), Step 5 Preview: 41.0 (#41)

BenchmarkGrok 4.20 Multi-AgentStep 5 Preview
LMArena Vision12591267

Multilingual Too close to call

Grok 4.20 Multi-Agent: 54.4 (#43), Step 5 Preview: 53.8 (#54)

BenchmarkGrok 4.20 Multi-AgentStep 5 Preview
LMArena Non-English14401432
LMArena Chinese14751519
LMArena Russian14571436
LMArena Spanish14471439
LMArena French1466—
LMArena German1456—
LMArena Japanese1405—
LMArena Korean1416—

Instruction Following Step 5 Preview leads

Grok 4.20 Multi-Agent: 74.8 (#84), Step 5 Preview: 76.0 (#50)

BenchmarkGrok 4.20 Multi-AgentStep 5 Preview
LMArena Instruction Following14201444

Long Context Too close to call

Grok 4.20 Multi-Agent: 43.7 (#88), Step 5 Preview: 44.5 (#64)

BenchmarkGrok 4.20 Multi-AgentStep 5 Preview
LMArena Longer Query14311455

Writing & Preference Grok 4.20 Multi-Agent leads

Grok 4.20 Multi-Agent: 64.0 (#59), Step 5 Preview: 62.8 (#71)

BenchmarkGrok 4.20 Multi-AgentStep 5 Preview
LMArena Text14501442
LMArena Creative Writing14361410
LMArena Multi-Turn14521447

Frequently asked questions

Is Grok 4.20 Multi-Agent better than Step 5 Preview?

Step 5 Preview is the stronger model overall, scoring 47.9 to 46.2 on the Noometry Index.

Which is cheaper, Grok 4.20 Multi-Agent or Step 5 Preview?

Step 5 Preview is cheaper. It lists at $1 per million input tokens and $2.70 per million output tokens; Grok 4.20 Multi-Agent lists at $1.25 and $2.50.

Is Grok 4.20 Multi-Agent 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 1M.

How many benchmarks do Grok 4.20 Multi-Agent and Step 5 Preview share?

14 benchmarks have published results for both models. Grok 4.20 Multi-Agent has 20 scored results on Noometry and Step 5 Preview has 18.

Related comparisons

Go deeper