Model comparison

Claude Haiku 5.5 vs Step 5 Preview

Claude Haiku 5.5 is the stronger model overall, scoring 49.5 to 47.9 on the Noometry Index.

Last verified . 1 shared benchmarks.

Claude Haiku 5.5 Anthropic

49.5

Rank #52 Confirmed

Step 5 Preview StepFun

47.9

Rank #58 Confirmed

Summary

  • They share 1 benchmark with published results for both. Claude Haiku 5.5 scores higher in 3 categories and Step 5 Preview in 2 categories; 5 gaps are clear of the uncertainty.
  • The widest gap is in math, where Claude Haiku 5.5 leads 73.6 to 46.1.
  • Claude Haiku 5.5 is cheaper at $0.10 / $0.50 per million input/output tokens, against $1 / $2.70 for Step 5 Preview.
  • Step 5 Preview accepts more context: 1.02M tokens versus 1M.

Side by side

Claude Haiku 5.5 and Step 5 Preview specifications
Claude Haiku 5.5Step 5 Preview
ProviderAnthropicStepFun
Noometry Index49.547.9
Released2026-10-072026-09-16
WeightsProprietaryProprietary
Context window1M1.02M
Max output128K66K
Input $ / M tokens$0.10$1
Output $ / M tokens$0.50$2.70
Results tracked918

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

Category by category

Coding Step 5 Preview leads

Claude Haiku 5.5: 49.2 (#50), Step 5 Preview: 51.8 (#36)

Coding benchmarks
BenchmarkClaude Haiku 5.5Step 5 Preview
LMArena WebDev15871564
SciCode—58.9%
LMArena Coding—1480

Reasoning Step 5 Preview leads

Claude Haiku 5.5: 35.2 (#69), Step 5 Preview: 40.0 (#57)

Reasoning benchmarks
BenchmarkClaude Haiku 5.5Step 5 Preview
NYT Connections (extended)65.7%—
CritPt—20.9%
LMArena Hard Prompts—1465
Mystery Game Puzzles30%—

Math Claude Haiku 5.5 leads

Claude Haiku 5.5: 73.6 (#18), Step 5 Preview: 46.1 (#72)

Math benchmarks
BenchmarkClaude Haiku 5.5Step 5 Preview
FrontierMath (Tiers 1-3)75.1%—
FrontierMath Tier 446.3%—
OTIS Mock AIME 2024-202598.9%—
ProofBench—42%
LMArena Math—1470

Knowledge Claude Haiku 5.5 leads

Claude Haiku 5.5: 50.8 (#70), Step 5 Preview: 41.2 (#112)

Knowledge benchmarks
BenchmarkClaude Haiku 5.5Step 5 Preview
GPQA Diamond89.6%—
SimpleQA Verified23.8%—
LMArena Expert—1470

Multimodal Claude Haiku 5.5 leads

Claude Haiku 5.5: 43.7 (#21), Step 5 Preview: 41.0 (#41)

Multimodal benchmarks
BenchmarkClaude Haiku 5.5Step 5 Preview
LMArena Vision—1267
Furniture Assembly47.5%—

Multilingual Not comparable

Claude Haiku 5.5: —, Step 5 Preview: 53.8 (#54)

Multilingual benchmarks
BenchmarkClaude Haiku 5.5Step 5 Preview
LMArena Non-English—1432
LMArena Chinese—1519
LMArena Russian—1436
LMArena Spanish—1439

Instruction Following Not comparable

Claude Haiku 5.5: —, Step 5 Preview: 76.0 (#50)

Instruction Following benchmarks
BenchmarkClaude Haiku 5.5Step 5 Preview
LMArena Instruction Following—1444

Long Context Not comparable

Claude Haiku 5.5: —, Step 5 Preview: 44.5 (#64)

Long Context benchmarks
BenchmarkClaude Haiku 5.5Step 5 Preview
LMArena Longer Query—1455

Writing & Preference Not comparable

Claude Haiku 5.5: —, Step 5 Preview: 62.8 (#71)

Writing & Preference benchmarks
BenchmarkClaude Haiku 5.5Step 5 Preview
LMArena Text—1442
LMArena Creative Writing—1410
LMArena Multi-Turn—1447

Frequently asked questions

Is Claude Haiku 5.5 better than Step 5 Preview?

Claude Haiku 5.5 is the stronger model overall, scoring 49.5 to 47.9 on the Noometry Index.

Which is cheaper, Claude Haiku 5.5 or Step 5 Preview?

Claude Haiku 5.5 is cheaper. It lists at $0.10 per million input tokens and $0.50 per million output tokens; Step 5 Preview lists at $1 and $2.70.

Is Claude Haiku 5.5 or Step 5 Preview better for coding?

Step 5 Preview scores higher on coding benchmarks: 51.8 versus 49.2 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 Claude Haiku 5.5 and Step 5 Preview share?

1 benchmark has published results for both models. Claude Haiku 5.5 has 9 scored results on Noometry and Step 5 Preview has 18.

Related comparisons

Go deeper