Skip to main content
BenchLM
Data

Claude Haiku 5.5 vs Claude Opus 4.6

Updated November 21, 2026. Rank says Claude Haiku 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 Haiku 5.5 has the higher public score estimate, 63.5 versus 59.9, but the 90% score intervals overlap. Treat that as a lead, not a settled winner. 7 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

Anthropic

63.5/100

Supported · Public rank #38

90% interval 59.5–67.6

Model B
Anthropic logo

Anthropic

59.9/100

Supported · Public rank #58

90% interval 55.9–64.0

API price per 1M tokens

Claude Haiku 5.5: $0.10 input / $0.50 output

Claude Opus 4.6: $5 input / $25 output

Context window

Claude Haiku 5.5: 1M

Claude Opus 4.6: 1M

Released

Claude Haiku 5.5: October 7, 2026

Claude Opus 4.6: February 1, 2026

Shared results
7
Claude Haiku 5.5 only
20
Claude Opus 4.6 only
32
Like-for-like categories
7 / 8
Supported: Claude Haiku 5.5 and Claude Opus 4.6How 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.

  • Coding work

    Code generation, repair, and software-engineering tasks

    Claude Haiku 5.5

    Claude Haiku 5.5 has the higher public coding point estimate, 67.9 to 61, with Supported evidence for both models. The conditional ranges do not establish rank confidence.

    Confidence: limited
  • Agentic work

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

    Claude Haiku 5.5

    Claude Haiku 5.5 has the higher public agentic point estimate, 68.3 to 58.8, with Supported evidence for both models. The conditional ranges do not establish rank confidence.

    Confidence: limited
  • Chat turn cost

    1K fresh input + 500 output tokens

    Claude Haiku 5.5

    Claude Haiku 5.5 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
  • Cache-heavy agent loop cost

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

    Claude Haiku 5.5

    Claude Haiku 5.5 has the lower estimated token cost for this stated workload. Claude Opus 4.6 has no published cached-input rate, so cached tokens use its listed input rate.

    Confidence: rate-fallback
  • Repository review cost

    50K fresh input + 3K output tokens

    Claude Haiku 5.5

    Claude Haiku 5.5 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates
  • 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.

67.9Claude Haiku 5.561.0Claude Opus 4.6

Like-for-like · BenchAlign v6.0

Claude Haiku 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.

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.

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 v6.0 lane in every category, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence. Directional and not-comparable rows remain visible, but they never receive a winner.

Agentic

Like-for-like
Claude Haiku 5.5
68.3
Supported · #25/182
Claude Opus 4.6
58.8
Supported · #62/182
Basis
BenchAlign v6.0 lane · 9 vs 11 public rows
Reading
Claude Haiku 5.5 leads · intervals overlap

Coding

Like-for-like
Claude Haiku 5.5
67.9
Supported · #28/180
Claude Opus 4.6
61.0
Supported · #47/180
Basis
BenchAlign v6.0 lane · 7 vs 10 public rows
Reading
Claude Haiku 5.5 leads · intervals overlap

Reasoning

Like-for-like
Claude Haiku 5.5
62.6
Supported · #38/161
Claude Opus 4.6
55.3
Supported · #59/161
Basis
BenchAlign v6.0 lane · 4 vs 1 public rows
Reading
Claude Haiku 5.5 leads · intervals overlap

Multimodal

Like-for-like
Claude Haiku 5.5
63.9
Supported · #39/119
Claude Opus 4.6
56.4
Supported · #59/119
Basis
BenchAlign v6.0 lane · 4 vs 5 public rows
Reading
Claude Haiku 5.5 leads · intervals overlap

Knowledge

Like-for-like
Claude Haiku 5.5
50.2
Supported · #84/158
Claude Opus 4.6
59.2
Supported · #50/158
Basis
BenchAlign v6.0 lane · 1 vs 9 public rows
Reading
Claude Opus 4.6 leads · intervals overlap

Instruction following

Like-for-like
Claude Haiku 5.5
58.3
Supported · #37/143
Claude Opus 4.6
56.9
Supported · #45/143
Basis
BenchAlign v6.0 lane · 0 vs 0 public rows
Reading
Claude Haiku 5.5 leads · intervals overlap

Math

Like-for-like
Claude Haiku 5.5
62.1
Supported · #29/125
Claude Opus 4.6
57.8
Supported · #40/125
Basis
BenchAlign v6.0 lane · 2 vs 3 public rows
Reading
Claude Haiku 5.5 leads · intervals overlap

Multilingual

Not comparable
Claude Haiku 5.5
56.1
Estimated · #36/88
Claude Opus 4.6
Not ranked
Basis
BenchAlign v6.0 lane · 0 vs 0 public rows
Reading
Not comparable

Ranks count the models scored in each category’s BenchAlign v6.0 lane, so denominators differ by category. 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 Haiku 5.5
$0.00035
Fits in one request
Claude Opus 4.6
$0.0175
Fits in one request

Claude Haiku 5.5 has the lower modeled cost

Costs use the listed standard API rates.

Repository review

50K fresh input + 3K output tokens

Claude Haiku 5.5
$0.0065
Fits in one request
Claude Opus 4.6
$0.325
Fits in one request

Claude Haiku 5.5 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 Haiku 5.5
$0.009
Fits in one request
Claude Opus 4.6
$1.35
Fits in one request
Cached input priced at the published list-input rate

Claude Haiku 5.5 has the lower modeled cost

Claude Opus 4.6 has no published cached-input rate, so cached tokens use its listed input rate.

Your workload

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

Claude Haiku 5.5
$2.00
per month
Claude Opus 4.6
$100.00
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

Price against score, among all priced models

Overall score against output price per 1M tokens, one model per family. Pricing as of November 21, 2026.

Claude Haiku 5.5Claude Opus 4.6

20406080100$0.10$0.50$1$5$10$50$100Output price, $ per 1M tokens (log)Overall score
  • Claude Haiku 5.5 is on the efficiency frontier: no cheaper model scores at least as high.
  • 30 models score at least as high as Claude Opus 4.6 for the same price or less. The nearest is Claude Opus 5 (score 75.1, also $25.00 per 1M output tokens).

Gray dots are the other priced models. The dark step line is the efficiency frontier: for a model under it, another model costs no more and scores at least as high.

See Claude Opus 4.6 on the full price vs performance chart

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 Haiku 5.5

$0.01 per 1M cached input tokens

Claude API pricing

Claude Opus 4.6

Not published

Reasoning profile

Claude Haiku 5.5

Reasoning

Claude Opus 4.6

Non-Reasoning

Weight access

Claude Haiku 5.5

Proprietary

Claude Opus 4.6

Proprietary

License

Claude Haiku 5.5

Proprietary

Claude Opus 4.6

Proprietary

Release date

Claude Haiku 5.5

2026-10-07

Claude Opus 4.6

2026-02-01

If you already use one of these models

Deployment change
Both entries list Anthropic as the provider. Confirm endpoint, model ID, limits, and feature support before switching.
Quality signal
Claude Haiku 5.5 has the higher public score estimate, 63.5 versus 59.9, but the 90% score intervals overlap.
Workload cost
Repository review: $0.0065 vs $0.325. Cache-heavy agent loop: $0.009 vs $1.35.
Context tradeoff
Both models list 1M.
Run the same representative tasks against both endpoints before changing production traffic.

Questions

Which is better, Claude Haiku 5.5 or Claude Opus 4.6?

Claude Haiku 5.5 has the higher public score estimate, 63.5 versus 59.9, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.

Which is better for coding, Claude Haiku 5.5 or Claude Opus 4.6?

Claude Haiku 5.5 has the higher public coding point estimate, 67.9 to 61, with Supported evidence for both models. The conditional ranges do not establish rank confidence.

Which is better for agentic tasks, Claude Haiku 5.5 or Claude Opus 4.6?

Claude Haiku 5.5 has the higher public agentic tasks point estimate, 68.3 to 58.8, with Supported evidence for both models. The conditional ranges do not establish rank confidence.

Which costs less, Claude Haiku 5.5 or Claude Opus 4.6?

For the stated presets, chat costs $0.00035 on Claude Haiku 5.5 and $0.0175 on Claude Opus 4.6; repository review costs $0.0065 and $0.325; the cache-heavy agent loop costs $0.009 and $1.35. Claude Opus 4.6 has no published cached-input rate, so cached tokens use its listed input rate.

Which has the larger context window, Claude Haiku 5.5 or Claude Opus 4.6?

Both models list the same context window, 1M.

Benchmark evidence

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

Browse raw public benchmark evidence59 rows

Agentic

  • Claude Haiku 5.558.62%
    Claude Opus 4.648.12%

    Claude Haiku 5.5 leads this result

  • Terminal-Bench 4.1

    Claude Haiku 5.540.00%
    Source
    Claude Opus 4.6—

    Not directly comparable

  • Agents' Last Exam (Linux)

    Claude Haiku 5.557.0%
    Source
    Claude Opus 4.6—

    Not directly comparable

  • Terminal-Bench 4.0

    Claude Haiku 5.539.20%
    Source
    Claude Opus 4.6—

    Not directly comparable

  • HLE w/ tools

    Claude Haiku 5.557.4%
    Source
    Claude Opus 4.6—

    Not directly comparable

  • MazeBench

    Claude Haiku 5.51%
    Source
    Claude Opus 4.6—

    Not directly comparable

  • Hermes Index

    Claude Haiku 5.537.0%
    Source
    Claude Opus 4.6—

    Not directly comparable

  • Hermes Bench

    Claude Haiku 5.571.0%
    Source
    Claude Opus 4.6—

    Not directly comparable

  • OSWorld 2.1 Offline

    Claude Haiku 5.572.4%
    Source
    Claude Opus 4.6—

    Not directly comparable

  • Terminal-Bench 2.0

    Claude Haiku 5.5—
    Claude Opus 4.665.4%
    Source

    Not directly comparable

  • BrowseComp

    Claude Haiku 5.5—
    Claude Opus 4.683.7%
    Source

    Not directly comparable

  • OSWorld-Verified

    Claude Haiku 5.5—
    Claude Opus 4.672.7%
    Source

    Not directly comparable

  • AndroidWorld

    Claude Haiku 5.5—
    Claude Opus 4.662.0%
    Source

    Not directly comparable

  • Claw-Eval

    Claude Haiku 5.5—
    Claude Opus 4.670.4%
    Source

    Not directly comparable

  • DeepSearchQA

    Claude Haiku 5.5—
    Claude Opus 4.673.7%
    Source

    Not directly comparable

  • CyberGym

    Claude Haiku 5.5—
    Claude Opus 4.666.6%
    Source

    Not directly comparable

  • ResearchClawBench

    Claude Haiku 5.5—
    Claude Opus 4.619.9%
    Source

    Not directly comparable

  • JobBench

    Claude Haiku 5.5—
    Claude Opus 4.636.7%
    Source

    Not directly comparable

  • ApprenticeBench

    Claude Haiku 5.5—
    Claude Opus 4.65%
    Source

    Not directly comparable

Coding

  • FrontierCode 1.1 Main

    Claude Haiku 5.546.4%
    Source
    Claude Opus 4.626.9%
    Source

    Claude Haiku 5.5 leads this result

  • Bug Hunt Bench

    Claude Haiku 5.521.5 fixes
    Source
    Claude Opus 4.6—

    Not directly comparable

  • FrontierSWE v2

    Claude Haiku 5.543.8%
    Source
    Claude Opus 4.6—

    Not directly comparable

  • FrontierCode 2.0

    Claude Haiku 5.528.9%
    Source
    Claude Opus 4.6—

    Not directly comparable

  • CursorBench 4.0

    Claude Haiku 5.548.4%
    Source
    Claude Opus 4.6—

    Not directly comparable

  • VulcanBench Frontier v4

    Claude Haiku 5.592.2%
    Source
    Claude Opus 4.6—

    Not directly comparable

  • CursorBench 4.1

    Claude Haiku 5.548.4%
    Source
    Claude Opus 4.6—

    Not directly comparable

  • SWE-bench Verified

    Claude Haiku 5.5—
    Claude Opus 4.680.8%
    Source

    Not directly comparable

  • SWE-bench Verified*

    Claude Haiku 5.5—
    Claude Opus 4.675.6%
    Source

    Not directly comparable

  • LiveCodeBench Pro

    Claude Haiku 5.5—
    Claude Opus 4.670.7%
    Source

    Not directly comparable

  • SWE-bench Pro

    Claude Haiku 5.5—
    Claude Opus 4.653.4%
    Source

    Not directly comparable

  • SWE-Rebench

    Claude Haiku 5.5—
    Claude Opus 4.665.3%
    Source

    Not directly comparable

  • React Native Evals

    Claude Haiku 5.5—
    Claude Opus 4.684.1%
    Source

    Not directly comparable

  • Terminal-Bench 2.0

    Claude Haiku 5.5—
    Claude Opus 4.665.4%
    Source

    Not directly comparable

  • Vibe Code Bench

    Claude Haiku 5.5—
    Claude Opus 4.657.57%
    Source

    Not directly comparable

  • PostTrainBench v1

    Claude Haiku 5.5—
    Claude Opus 4.623.2%
    Source

    Not directly comparable

Reasoning

  • ARC-AGI-1

    Claude Haiku 5.589.00%
    Source
    Claude Opus 4.6—

    Not directly comparable

  • ARC-AGI-2

    Claude Haiku 5.561.3%
    Source
    Claude Opus 4.6—

    Not directly comparable

  • ARC-AGI-3

    Claude Haiku 5.50.5%
    Source
    Claude Opus 4.60.5%
    Source

    Claude Opus 4.6 leads this result

  • ARC-AGI-3 Provider Adapter

    Claude Haiku 5.52.82%
    Source
    Claude Opus 4.6—

    Not directly comparable

Multimodal

  • ParseBench-2

    Claude Haiku 5.571.24%
    Source
    Claude Opus 4.6—

    Not directly comparable

  • PerceptionBench (tools)

    Claude Haiku 5.578.6%
    Source
    Claude Opus 4.6—

    Not directly comparable

  • ParseBench-1

    Shared source
    Claude Haiku 5.556.06%
    Claude Opus 4.653.80%

    Claude Haiku 5.5 leads this result

  • Chartography (no tools)

    Claude Haiku 5.546.4%
    Source
    Claude Opus 4.6—

    Not directly comparable

  • MMMU-Pro

    Claude Haiku 5.5—
    Claude Opus 4.677.3%
    Source

    Not directly comparable

  • ERQA

    Claude Haiku 5.5—
    Claude Opus 4.651.6%
    Source

    Not directly comparable

  • ScreenSpot Pro

    Claude Haiku 5.5—
    Claude Opus 4.683.1%
    Source

    Not directly comparable

  • MedXpertQA (MM)

    Claude Haiku 5.5—
    Claude Opus 4.664.8%
    Source

    Not directly comparable

Knowledge

  • HLE w/o tools

    Claude Haiku 5.545.9%
    Source
    Claude Opus 4.640%
    Source

    Claude Haiku 5.5 leads this result

  • GPQA

    Claude Haiku 5.5—
    Claude Opus 4.691.3%
    Source

    Not directly comparable

  • GPQA-D

    Claude Haiku 5.5—
    Claude Opus 4.689.2%
    Source

    Not directly comparable

  • SuperGPQA

    Claude Haiku 5.5—
    Claude Opus 4.695%
    Source

    Not directly comparable

  • MMLU-Pro

    Claude Haiku 5.5—
    Claude Opus 4.682%
    Source

    Not directly comparable

  • MMLU-Pro (Arcee)

    Claude Haiku 5.5—
    Claude Opus 4.689.1%
    Source

    Not directly comparable

  • HLE

    Claude Haiku 5.5—
    Claude Opus 4.653%
    Source

    Not directly comparable

  • HealthBench Hard

    Claude Haiku 5.5—
    Claude Opus 4.614.8%
    Source

    Not directly comparable

  • MedXpertQA (Text)

    Claude Haiku 5.5—
    Claude Opus 4.652.1%
    Source

    Not directly comparable

Math

  • FrontierMath v2 (Tiers 1-3)

    Shared source
    Claude Haiku 5.575.088%
    Claude Opus 4.665.965%

    Claude Haiku 5.5 leads this result

  • FrontierMath v2 (Tier 4)

    Shared source
    Claude Haiku 5.546.341%
    Claude Opus 4.626.829%

    Claude Haiku 5.5 leads this result

  • AIME25 (Arcee)

    Claude Haiku 5.5—
    Claude Opus 4.699.8%
    Source

    Not directly comparable

59 public results · 7 shared

Watch Claude Haiku 5.5 vs Claude Opus 4.6

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 21, 2026