Skip to main content
BenchLM
Data

Claude Sonnet 5.5 vs Step 5 Preview

Updated November 12, 2026. Rank says Claude Sonnet 5.5 is ahead. Price, access, and your workload can each overturn that. Public scores include evidence status and uncertainty.

Share or export
Share on XLinkedInAPI/MCP

Decision reading

Claude Sonnet 5.5 has the higher public point estimate, 83.07 versus 55.48. Their conditional score ranges do not overlap. These ranges do not establish rank confidence. 8 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.

Model A
Anthropic logo
Claude Sonnet 5.5

Anthropic

83.07/100

Supported · Public rank #6

90% interval 78.8–87.3

Model B
StepFun logo
Step 5 Preview

StepFun

55.48/100

Estimated · Public rank #66

Conditional range 45.8–65.2

API price per 1M tokens

Claude Sonnet 5.5: $2 input / $10 output

Step 5 Preview: $1 input / $2.70 output

Context window

Claude Sonnet 5.5: 1M

Step 5 Preview: 1M

Released

Claude Sonnet 5.5: September 28, 2026

Step 5 Preview: September 20, 2026

Shared results
8
Claude Sonnet 5.5 only
56
Step 5 Preview only
16
Like-for-like categories
1 / 8
Supported: Claude Sonnet 5.5 · Estimated: Step 5 Preview. Conditional ranges do not establish rank confidence.How the comparison works

Which one for your work

Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.

  • Chat turn cost

    1K fresh input + 500 output tokens

    Step 5 Preview

    Step 5 Preview has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates

  • Cache-heavy agent loop cost

    200K cached + 20K fresh input + 10K output tokens

    Step 5 Preview

    Step 5 Preview has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates
Show secondary and unsupported calls

  • Repository review cost

    50K fresh input + 3K output tokens

    Step 5 Preview

    Step 5 Preview has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates

  • Coding work

    Code generation, repair, and software-engineering tasks

    Not enough matched evidence

    Step 5 Preview is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like.

    Confidence: limited

  • Agentic work

    Tool use, computer use, and multi-step task completion

    Not enough matched evidence

    Step 5 Preview is scored on Estimated evidence for agentic, so the reading is directional rather than like-for-like.

    Confidence: limited

  • Long documents

    Prompts that approach the documented context limit

    No clear pick

    The documented context windows are equal.

    Confidence: documented

Which one for a specific job

Choose a job from the LLM Selector's task catalog to see the category row it rests on, under the same basis rules as the table below. A directional row stays directional; choosing a job never creates a winner.

The same task catalog as the LLM Selector. Each job names the evidence surface it rests on; nothing here adds a new score.

81.0Claude Sonnet 5.549.1Step 5 Preview

Directional only · BenchAlign v5.8

Claude Sonnet 5.5 has the higher coding point estimate. Conditional score ranges do not establish rank confidence.

Coding scores combine specific tasks and setups. Match the editor, harness, and effort to your workflow.

Open the coding leaderboardShortlist models for this job
Same basis rules as the category table below

Or describe your own job

The LLM Selector reads what you need a model for and returns a shortlist, ranked in the same order the job pages publish.

Describe the task and constraints. This selector recommends models; it cannot carry out the task or access your files and accounts.

We keep what you type here, drafts included, for 180 days, and after that only the job and requirements read from it. Nothing is tied to your IP address or account. Leave out names and other personal details. Privacy

What is actually comparable

Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.

4 categories rest on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.

Shape of the matched evidence

Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.

Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.

Bars run 0–100 on each benchmark’s normalized display scale

Category results, on a stated basis

Each row shows the public-lane category score for both models: the BenchAlign v5.8 lane for agentic, coding, reasoning, multimodal, and knowledge, and the weighted public rows elsewhere, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence or the same weighted set. Directional and not-comparable rows remain visible, but they never receive a winner.

Knowledge

Like-for-like
Claude Sonnet 5.5
82.1
Supported · #6/180
Step 5 Preview
61.2
Supported · #45/180
Basis
BenchAlign v5.8 lane · 17 vs 2 public rows
Reading
Claude Sonnet 5.5 leads

Agentic

Directional only
Claude Sonnet 5.5
72.7
Supported · #9/135
Step 5 Preview
54.2
Estimated · #51/135
Basis
BenchAlign v5.8 lane · 19 vs 13 public rows
Reading
Directional only

Coding

Directional only
Claude Sonnet 5.5
81.0
Supported · #4/162
Step 5 Preview
49.1
Estimated · #52/162
Basis
BenchAlign v5.8 lane · 12 vs 6 public rows
Reading
Directional only

Reasoning

Directional only
Claude Sonnet 5.5
77.0
Supported · #6/160
Step 5 Preview
53.2
Estimated · #49/160
Basis
BenchAlign v5.8 lane · 0 vs 1 public rows
Reading
Directional only

Multimodal

Directional only
Claude Sonnet 5.5
78.9
Estimated · #7/67
Step 5 Preview
60.2
Estimated · #37/67
Basis
BenchAlign v5.8 lane · 10 vs 2 public rows
Reading
Directional only

Multilingual

Not comparable
Claude Sonnet 5.5
Not ranked
Step 5 Preview
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Instruction following

Not comparable
Claude Sonnet 5.5
Not ranked
Step 5 Preview
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Math

Not comparable
Claude Sonnet 5.5
89.7
Unranked · 4 rankable rows
Step 5 Preview
Not ranked
Basis
Provisional lane · 2 vs 0 weighted rows
Reading
Not comparable

Ranks count the models scored in each category’s lane, so the agentic, coding, reasoning, multimodal, and knowledge denominators (BenchAlign v5.8) differ from the provisional-lane categories. Unranked scores sit on the provisional lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.

Supported evidence per lane · bars run 0–100Methodology

What each workload costs

Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.

Chat turn

1K fresh input + 500 output tokens

Claude Sonnet 5.5
$0.007
Fits in one request
Step 5 Preview
$0.00235
Fits in one request

Step 5 Preview has the lower modeled cost

Costs use the listed standard API rates.

Repository review

50K fresh input + 3K output tokens

Claude Sonnet 5.5
$0.13
Fits in one request
Step 5 Preview
$0.0581
Fits in one request

Step 5 Preview has the lower modeled cost

Costs use the listed standard API rates.

Cache-heavy agent loop

200K cached + 20K fresh input + 10K output tokens

Claude Sonnet 5.5
$0.16
Fits in one request
Step 5 Preview
$0.057
Fits in one request

Step 5 Preview has the lower modeled cost

Costs use the listed standard API rates.

Your workload

Enter a month of tokens to price it at both models’ listed API rates.

Claude Sonnet 5.5
$40.00
per month
Step 5 Preview
$15.40
per month

Standard listed rates, no batch or volume discounts.

Cached input falls back to the list input rate only where a cached rate is unpublished

Specification differences

Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.

Cached-input rate

A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.

Claude Sonnet 5.5

$0.1 per 1M cached input tokens

Claude API pricing

Step 5 Preview

$0.05 per 1M cached input tokens

StepFun pricing and rate limits

Reasoning profile

Claude Sonnet 5.5

Reasoning

Step 5 Preview

Reasoning

Weight access

Claude Sonnet 5.5

Proprietary

Step 5 Preview

Open Weight

License

Claude Sonnet 5.5

Proprietary

Step 5 Preview

Open Weight

Release date

Claude Sonnet 5.5

2026-09-28

Step 5 Preview

2026-09-20

If you already use one of these models

Deployment change
The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
Quality signal
Claude Sonnet 5.5 has the higher public point estimate, 83.07 versus 55.48. Their conditional score ranges do not overlap. These ranges do not establish rank confidence.
Workload cost
Repository review: $0.13 vs $0.0581. Cache-heavy agent loop: $0.16 vs $0.057.
Context tradeoff
Both models list 1M.
Run the same representative tasks against both endpoints before changing production traffic.

Questions

Which is better, Claude Sonnet 5.5 or Step 5 Preview?

Claude Sonnet 5.5 has the higher public point estimate, 83.07 versus 55.48. Their conditional score ranges do not overlap. These ranges do not establish rank confidence. The page therefore keeps the decision tied to the specific documented workload.

Which is better for coding, Claude Sonnet 5.5 or Step 5 Preview?

Claude Sonnet 5.5 scores higher for coding on the public lane, 81 to 49.1. Step 5 Preview is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like. BenchLM does not name a winner for coding from a directional reading; read the shared benchmark rows directly and test the models on the same task set.

Which is better for agentic tasks, Claude Sonnet 5.5 or Step 5 Preview?

Claude Sonnet 5.5 scores higher for agentic tasks on the public lane, 72.7 to 54.2. Step 5 Preview is scored on Estimated evidence for agentic tasks, so the reading is directional rather than like-for-like. BenchLM does not name a winner for agentic tasks from a directional reading; read the shared benchmark rows directly and test the models on the same task set.

Which costs less, Claude Sonnet 5.5 or Step 5 Preview?

For the stated presets, chat costs $0.007 on Claude Sonnet 5.5 and $0.00235 on Step 5 Preview; repository review costs $0.13 and $0.0581; the cache-heavy agent loop costs $0.16 and $0.057. Costs use the listed standard API rates.

Which has the larger context window, Claude Sonnet 5.5 or Step 5 Preview?

Both models list the same context window, 1M.

Self-host vs API cost

Estimates at 50,000 req/day · 1000 tokens/req average.

Claude Sonnet 5.5
API / mo$9,000
Self-host / moNot listed
Break-even—
Proprietary model — self-hosting not applicable.
Step 5 Preview
API / mo$2,775
Self-host / mo$27,962
Break-evenNot reached at 70% load
Model the full break-even

Benchmark evidence

The full public result ledger is available for audit without forcing a wide desktop table onto a phone.

Browse raw public benchmark evidence80 rows

Agentic

  • Terminal-Bench 4.1

    Claude Sonnet 5.563.90%
    Source
    Step 5 Preview—

    Not directly comparable

  • Terminal-Bench 4.0

    Claude Sonnet 5.570.60%
    Source
    Step 5 Preview33.30%
    Source

    Claude Sonnet 5.5 leads this result

  • HLE w/ tools

    Claude Sonnet 5.564.5%
    Source
    Step 5 Preview—

    Not directly comparable

  • DRACO

    Claude Sonnet 5.587.0%
    Source
    Step 5 Preview83.3%
    Source

    Claude Sonnet 5.5 leads this result

  • Terminal-Bench-Science 0.1

    Claude Sonnet 5.559.9%
    Source
    Step 5 Preview—

    Not directly comparable

  • LAB all-pass (Harvey held-out)

    Claude Sonnet 5.510.0%
    Source
    Step 5 Preview—

    Not directly comparable

  • LAB criterion-pass (Harvey held-out)

    Claude Sonnet 5.593.1%
    Source
    Step 5 Preview—

    Not directly comparable

  • Toolathlon Verified Pass@3

    Claude Sonnet 5.585.2%
    Source
    Step 5 Preview—

    Not directly comparable

  • Toolathlon Verified Pass³

    Claude Sonnet 5.568.5%
    Source
    Step 5 Preview—

    Not directly comparable

  • Toolathlon Verified avg. turns

    Claude Sonnet 5.531.6 turns
    Source
    Step 5 Preview—

    Not directly comparable

  • AutomationBench (Zapier 1.0.6)

    Claude Sonnet 5.544.70%
    Source
    Step 5 Preview—

    Not directly comparable

  • Toolathlon-Verified

    Claude Sonnet 5.577.8%
    Source
    Step 5 Preview74.1%
    Source

    Claude Sonnet 5.5 leads this result

  • Hermes Index

    Shared source
    Claude Sonnet 5.553.1%
    Step 5 Preview32.8%

    Claude Sonnet 5.5 leads this result

  • Hermes Bench

    Shared source
    Claude Sonnet 5.573.5%
    Step 5 Preview69.2%

    Claude Sonnet 5.5 leads this result

  • MazeBench

    Claude Sonnet 5.58%
    Source
    Step 5 Preview—

    Not directly comparable

  • OSWorld 2.1

    Claude Sonnet 5.580.1%
    Source
    Step 5 Preview—

    Not directly comparable

  • OSWorld 2.1 Offline

    Claude Sonnet 5.583.9%
    Source
    Step 5 Preview—

    Not directly comparable

  • Binary Exploitation Benchmark

    Claude Sonnet 5.548
    Source
    Step 5 Preview—

    Not directly comparable

  • Terminal-Bench 2.1 (Vals)

    Claude Sonnet 5.583.1%
    Source
    Step 5 Preview—

    Not directly comparable

  • Terminal-Bench 2.1

    Claude Sonnet 5.5—
    Step 5 Preview85.0%
    Source

    Not directly comparable

  • CyberGym

    Claude Sonnet 5.5—
    Step 5 Preview84.7%
    Source

    Not directly comparable

  • AutomationBench

    Claude Sonnet 5.5—
    Step 5 Preview44.0%
    Source

    Not directly comparable

  • MCP Atlas

    Claude Sonnet 5.5—
    Step 5 Preview85.6%
    Source

    Not directly comparable

  • JobBench

    Claude Sonnet 5.5—
    Step 5 Preview59.0%
    Source

    Not directly comparable

  • APEX-Agents

    Claude Sonnet 5.5—
    Step 5 Preview37.8%
    Source

    Not directly comparable

  • BrowseComp

    Claude Sonnet 5.5—
    Step 5 Preview88.7%
    Source

    Not directly comparable

  • Agents' Last Exam

    Claude Sonnet 5.5—
    Step 5 Preview29.5%
    Source

    Not directly comparable

Coding

  • FrontierCode 1.1 Main

    Claude Sonnet 5.546.2%
    Source
    Step 5 Preview—

    Not directly comparable

  • CursorBench 4.0

    Claude Sonnet 5.555.5%
    Source
    Step 5 Preview—

    Not directly comparable

  • SWE-bench Pro

    Claude Sonnet 5.581.3%
    Source
    Step 5 Preview—

    Not directly comparable

  • SWE Multilingual

    Claude Sonnet 5.590.3%
    Source
    Step 5 Preview—

    Not directly comparable

  • SWE Multimodal

    Claude Sonnet 5.554.3%
    Source
    Step 5 Preview—

    Not directly comparable

  • DeepSWE

    Claude Sonnet 5.571.0%
    Source
    Step 5 Preview67.7%
    Source

    Claude Sonnet 5.5 leads this result

  • FrontierCode 1.1 Extended

    Claude Sonnet 5.559.1%
    Source
    Step 5 Preview—

    Not directly comparable

  • ProgramBench

    Claude Sonnet 5.579.7%
    Source
    Step 5 Preview80.5%
    Source

    Step 5 Preview leads this result

  • FrontierSWE v2

    Claude Sonnet 5.561.9%
    Source
    Step 5 Preview—

    Not directly comparable

  • Bug Hunt Bench

    Claude Sonnet 5.551.3 fixes
    Source
    Step 5 Preview—

    Not directly comparable

  • FrontierCode 2.0

    Claude Sonnet 5.541.0%
    Source
    Step 5 Preview—

    Not directly comparable

  • VulcanBench Frontier v4

    Claude Sonnet 5.592.0%
    Source
    Step 5 Preview—

    Not directly comparable

  • Terminal-Bench 2.1

    Claude Sonnet 5.5—
    Step 5 Preview85.0%
    Source

    Not directly comparable

  • SciCode

    Claude Sonnet 5.5—
    Step 5 Preview58.9%
    Source

    Not directly comparable

  • sweMarathon

    Claude Sonnet 5.5—
    Step 5 Preview72.7%
    Source

    Not directly comparable

  • MLS-Bench Lite

    Claude Sonnet 5.5—
    Step 5 Preview40.5%
    Source

    Not directly comparable

Reasoning

  • CritPt

    Claude Sonnet 5.5—
    Step 5 Preview20.9%
    Source

    Not directly comparable

Multimodal

  • ExtractBench Pro

    Claude Sonnet 5.532%
    Source
    Step 5 Preview—

    Not directly comparable

  • ExtractBench (LlamaIndex)

    Claude Sonnet 5.594.68%
    Source
    Step 5 Preview—

    Not directly comparable

  • ParseBench

    Claude Sonnet 5.569.21%
    Source
    Step 5 Preview—

    Not directly comparable

  • Chartography (no tools)

    Claude Sonnet 5.561.6%
    Source
    Step 5 Preview—

    Not directly comparable

  • Chartography (tools)

    Claude Sonnet 5.590.2%
    Source
    Step 5 Preview—

    Not directly comparable

  • BenchCAD Vision2Code (no tools)

    Claude Sonnet 5.50.747
    Source
    Step 5 Preview—

    Not directly comparable

  • BenchCAD Vision2Code (tools)

    Claude Sonnet 5.50.963
    Source
    Step 5 Preview—

    Not directly comparable

  • Biomedical image analysis

    Claude Sonnet 5.572.2%
    Source
    Step 5 Preview—

    Not directly comparable

  • OfficeQA

    Claude Sonnet 5.576.9%
    Source
    Step 5 Preview—

    Not directly comparable

  • OfficeQA Pro

    Claude Sonnet 5.565.6%
    Source
    Step 5 Preview60.3%
    Source

    Claude Sonnet 5.5 leads this result

  • MMMU-Pro

    Claude Sonnet 5.5—
    Step 5 Preview76%
    Source

    Not directly comparable

Knowledge

  • HealthBench (raw)

    Claude Sonnet 5.569.4%
    Source
    Step 5 Preview—

    Not directly comparable

  • HealthBench (length-adjusted)

    Claude Sonnet 5.565.4%
    Source
    Step 5 Preview—

    Not directly comparable

  • HealthBench Professional (raw)

    Claude Sonnet 5.577.1%
    Source
    Step 5 Preview—

    Not directly comparable

  • HealthBench Professional

    Claude Sonnet 5.569.2%
    Source
    Step 5 Preview—

    Not directly comparable

  • BioMysteryBench (human-solvable)

    Claude Sonnet 5.589.2%
    Source
    Step 5 Preview—

    Not directly comparable

  • BioMysteryBench (human-difficult)

    Claude Sonnet 5.544.7%
    Source
    Step 5 Preview—

    Not directly comparable

  • SpatialBench Verified

    Claude Sonnet 5.572.5%
    Source
    Step 5 Preview—

    Not directly comparable

  • SingleCellBench

    Claude Sonnet 5.559.1%
    Source
    Step 5 Preview—

    Not directly comparable

  • Protein Design

    Claude Sonnet 5.551.0%
    Source
    Step 5 Preview—

    Not directly comparable

  • Morphology-to-molecule matching

    Claude Sonnet 5.525.0%
    Source
    Step 5 Preview—

    Not directly comparable

  • Medicinal chemistry

    Claude Sonnet 5.565.3%
    Source
    Step 5 Preview—

    Not directly comparable

  • Protein Design library ranking

    Claude Sonnet 5.554.8%
    Source
    Step 5 Preview—

    Not directly comparable

  • De novo protein-binder design

    Claude Sonnet 5.582.3%
    Source
    Step 5 Preview—

    Not directly comparable

  • Protocols (troubleshooting)

    Claude Sonnet 5.567.3%
    Source
    Step 5 Preview—

    Not directly comparable

  • Protocols (understanding)

    Claude Sonnet 5.566.6%
    Source
    Step 5 Preview—

    Not directly comparable

  • HLE w/o tools

    Claude Sonnet 5.556.9%
    Source
    Step 5 Preview—

    Not directly comparable

  • PhysicianBench

    Claude Sonnet 5.563.2%
    Source
    Step 5 Preview—

    Not directly comparable

  • GPQA-D

    Claude Sonnet 5.5—
    Step 5 Preview93.5%
    Source

    Not directly comparable

  • HLE

    Claude Sonnet 5.5—
    Step 5 Preview46.5%
    Source

    Not directly comparable

Multilingual

  • GMMLU

    Claude Sonnet 5.592.1%
    Source
    Step 5 Preview—

    Not directly comparable

  • MILU

    Claude Sonnet 5.591.6%
    Source
    Step 5 Preview—

    Not directly comparable

Math

  • ArXivMath Aug. 2026 (no tools)

    Claude Sonnet 5.586.8%
    Source
    Step 5 Preview—

    Not directly comparable

  • ArXivMath Aug. 2026 (tools)

    Claude Sonnet 5.595.2%
    Source
    Step 5 Preview—

    Not directly comparable

  • FrontierMath v2 (Tiers 1-3)

    Claude Sonnet 5.588.772%
    Source
    Step 5 Preview—

    Not directly comparable

  • FrontierMath v2 (Tier 4)

    Claude Sonnet 5.580.488%
    Source
    Step 5 Preview—

    Not directly comparable

80 public results · 8 shared

Related comparisons

Compare API pricingRead the methodologyOpen the model selector

Watch Claude Sonnet 5.5 vs Step 5 Preview

One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.

Read a sample issue

Join 7,000+ subscribers.

Last updated November 12, 2026