Skip to main content
Radar

Five or fewer confirmed AI changes, with original sources, on mornings when something changed.A free source-linked morning brief.

See the free Radar Brief
BenchLM
Anthropic logo
Model A
Claude Opus 5

Anthropic

80.66/100

Supported · Public rank #4

90% interval 78.3–83.0

Claude Opus 5 vs A.X series

Updated September 4, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.

Model B
A.X series

SK Telecom

—

Evidence status unavailable

90% interval unavailable

Decision reading

The public evidence has no benchmark result shared by both models, so it does not support a quality verdict. Use the documented cost, context, and runtime rows instead.

0 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.

Share or export

Share on XLinkedInSocial cardCSVJSON

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.

  • Long documents

    Prompts that approach the documented context limit

    Claude Opus 5

    Claude Opus 5 has the larger documented context window.

    Confidence: documented

Show secondary and unsupported calls

  • Coding work

    Code generation, repair, and software-engineering tasks

    Not enough matched evidence

    A.X series is not ranked on the public lane for coding, so no winner is named for coding.

    Confidence: limited

  • Agentic work

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

    Not enough matched evidence

    A.X series is not ranked on the public lane for agentic, so no winner is named for agentic.

    Confidence: limited

  • Chat turn cost

    1K fresh input + 500 output tokens

    Not enough matched evidence

    A complete comparable API-rate estimate is not available for both models.

    Confidence: listed-rates

  • Cache-heavy agent loop cost

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

    Not enough matched evidence

    The page does not recommend a cost winner because at least one model cannot fit the stated workload in one request. A.X series does not fit this workload in one request. A.X series has no comparable published API token rate.

    Confidence: listed-rates

  • Repository review cost

    50K fresh input + 3K output tokens

    Not enough matched evidence

    A complete comparable API-rate estimate is not available for both models.

    Confidence: listed-rates

What is actually comparable

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

Shared results
0
Claude Opus 5 only
73
A.X series only
0
Like-for-like categories
0 / 8

Category results, on a stated basis

Each row shows the public-lane category score for both models: the BenchAlign lane for agentic, coding, 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.

Agentic

Not comparable
Claude Opus 5
77.4
Supported · #2/151
A.X series
Not ranked
Basis
BenchAlign lane · 19 vs 0 public rows
Reading
Not comparable

Coding

Not comparable
Claude Opus 5
75.6
Supported · #3/183
A.X series
Not ranked
Basis
BenchAlign lane · 16 vs 0 public rows
Reading
Not comparable

Reasoning

Not comparable
Claude Opus 5
75.6
#13/22
A.X series
Not ranked
Basis
Provisional lane · 2 vs 0 weighted rows
Reading
Not comparable

Knowledge

Not comparable
Claude Opus 5
82.1
Supported · #3/181
A.X series
Not ranked
Basis
BenchAlign lane · 19 vs 0 public rows
Reading
Not comparable

Math

Not comparable
Claude Opus 5
Not ranked
A.X series
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Multilingual

Not comparable
Claude Opus 5
Not ranked
A.X series
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Multimodal

Not comparable
Claude Opus 5
88.7
#2/48
A.X series
Not ranked
Basis
Provisional lane · 1 vs 0 weighted rows
Reading
Not comparable

Instruction following

Not comparable
Claude Opus 5
Not ranked
A.X series
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

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

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.

A shared-evidence shape is not available.

BenchLM does not draw a radar or infer missing axes when the matched evidence is too sparse.

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 Opus 5
$0.0175
Fits in one request
A.X series
API rate not published
Fits in one request

A.X series has no comparable published API token rate.

Repository review

50K fresh input + 3K output tokens

Claude Opus 5
$0.325
Fits in one request
A.X series
API rate not published
Fits in one request

A.X series has no comparable published API token rate.

Cache-heavy agent loop

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

Claude Opus 5
$0.45
Fits in one request
A.X series
API rate not published
Does not fit in one request
Cached-input rate unavailable

A.X series does not fit this workload in one request. A.X series has no comparable published API token rate.

Specification differences

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

Context window

Maximum documented context; output-token limits may be lower.

A.X series

64K

Cached-input rate

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

Claude Opus 5

$0.5 per 1M cached input tokens

Claude API pricing

A.X series

No comparable hosted API rate

Reasoning profile

Claude Opus 5

Reasoning

A.X series

Non-Reasoning

Weight access

Claude Opus 5

Proprietary

A.X series

Proprietary

License

Claude Opus 5

Proprietary

A.X series

Proprietary

Release date

Claude Opus 5

2026-07-24

A.X series

Not sourced

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
The public evidence has no benchmark result shared by both models, so it does not support a quality verdict.
Workload cost
A complete comparable API-rate estimate is not available for both models.
Context tradeoff
Claude Opus 5 has the larger documented window (1M).

Run the same representative tasks against both endpoints before changing production traffic.

Benchmark evidence

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

Browse raw public benchmark evidence73 rows

Agentic

  • Terminal-Bench 3.0

    Claude Opus 542.7%
    Source
    A.X series—

    Not directly comparable

  • BrowseComp

    Claude Opus 590.8%
    Source
    A.X series—

    Not directly comparable

  • HLE w/ tools

    Claude Opus 564.7%
    Source
    A.X series—

    Not directly comparable

  • DeepSearchQA

    Claude Opus 595.0%
    Source
    A.X series—

    Not directly comparable

  • DRACO

    Claude Opus 588.6%
    Source
    A.X series—

    Not directly comparable

  • BrowseComp (10-agent, prerelease)

    Claude Opus 593.6%
    Source
    A.X series—

    Not directly comparable

  • OSWorld 2.0

    Claude Opus 570.6%
    Source
    A.X series—

    Not directly comparable

  • MCP Atlas

    Claude Opus 585.8%
    Source
    A.X series—

    Not directly comparable

  • MCP-Atlas claim coverage

    Claude Opus 589.1%
    Source
    A.X series—

    Not directly comparable

  • LAB all-pass (Anthropic harness)

    Claude Opus 523.58%
    Source
    A.X series—

    Not directly comparable

  • LAB criterion-pass (Anthropic harness)

    Claude Opus 593.74%
    Source
    A.X series—

    Not directly comparable

  • LAB all-pass (Harvey held-out)

    Claude Opus 511.7%
    Source
    A.X series—

    Not directly comparable

  • LAB criterion-pass (Harvey held-out)

    Claude Opus 594.1%
    Source
    A.X series—

    Not directly comparable

  • Toolathlon-Verified

    Claude Opus 580.6%
    Source
    A.X series—

    Not directly comparable

  • Toolathlon Verified Pass@3

    Claude Opus 587.0%
    Source
    A.X series—

    Not directly comparable

  • Toolathlon Verified Pass³

    Claude Opus 573.1%
    Source
    A.X series—

    Not directly comparable

  • Toolathlon Verified avg. turns

    Claude Opus 523.5 turns
    Source
    A.X series—

    Not directly comparable

  • AutomationBench

    Claude Opus 526.0%
    Source
    A.X series—

    Not directly comparable

  • Terminal-Bench 2.1 (Vals)

    Claude Opus 584.6%
    Source
    A.X series—

    Not directly comparable

Coding

  • Bug Hunt Bench

    Claude Opus 527 fixes
    Source
    A.X series—

    Not directly comparable

  • SWE-bench Verified

    Claude Opus 596%
    Source
    A.X series—

    Not directly comparable

  • SWE-bench Pro

    Claude Opus 579.2%
    Source
    A.X series—

    Not directly comparable

  • SWE Multilingual

    Claude Opus 589.5%
    Source
    A.X series—

    Not directly comparable

  • SWE Multimodal

    Claude Opus 559.4%
    Source
    A.X series—

    Not directly comparable

  • deepSwe

    Claude Opus 568.8%
    Source
    A.X series—

    Not directly comparable

  • FrontierCode 1.1 Main

    Claude Opus 553.4%
    Source
    A.X series—

    Not directly comparable

  • FrontierCode 1.1 Extended

    Claude Opus 563.6%
    Source
    A.X series—

    Not directly comparable

  • FrontierSWE v2

    Claude Opus 552.0%
    Source
    A.X series—

    Not directly comparable

  • ProgramBench (episode 1)

    Claude Opus 583.0%
    Source
    A.X series—

    Not directly comparable

  • ProgramBench

    Claude Opus 593.0%
    Source
    A.X series—

    Not directly comparable

  • cursorBench32

    Claude Opus 570.0%
    Source
    A.X series—

    Not directly comparable

  • VulcanBench v3

    Claude Opus 587.0%
    Source
    A.X series—

    Not directly comparable

  • VulcanBench CII v1

    Claude Opus 596.4%
    Source
    A.X series—

    Not directly comparable

  • LiveCodeBench (Vals)

    Claude Opus 589.0%
    Source
    A.X series—

    Not directly comparable

  • SWE-bench (Vals)

    Claude Opus 597.0%
    Source
    A.X series—

    Not directly comparable

Reasoning

  • ARC-AGI-1

    Claude Opus 597.50%
    Source
    A.X series—

    Not directly comparable

  • ARC-AGI-2

    Claude Opus 590.4%
    Source
    A.X series—

    Not directly comparable

  • ARC-AGI-3

    Claude Opus 530.2%
    Source
    A.X series—

    Not directly comparable

Knowledge

  • HLE

    Claude Opus 564.7%
    Source
    A.X series—

    Not directly comparable

  • HLE w/o tools

    Claude Opus 556.3%
    Source
    A.X series—

    Not directly comparable

  • HLE-Verified

    Claude Opus 554.4%
    Source
    A.X series—

    Not directly comparable

  • LABBench2

    Claude Opus 584.2%
    Source
    A.X series—

    Not directly comparable

  • HealthBench (raw)

    Claude Opus 567.1%
    Source
    A.X series—

    Not directly comparable

  • HealthBench (length-adjusted)

    Claude Opus 557.8%
    Source
    A.X series—

    Not directly comparable

  • HealthBench Professional

    Claude Opus 559.8%
    Source
    A.X series—

    Not directly comparable

  • HealthBench Professional (raw)

    Claude Opus 573.4%
    Source
    A.X series—

    Not directly comparable

  • BioMysteryBench (human-solvable)

    Claude Opus 590.1%
    Source
    A.X series—

    Not directly comparable

  • BioMysteryBench (human-difficult)

    Claude Opus 549.4%
    Source
    A.X series—

    Not directly comparable

  • SpatialBench Verified

    Claude Opus 572.5%
    Source
    A.X series—

    Not directly comparable

  • SingleCellBench

    Claude Opus 560.6%
    Source
    A.X series—

    Not directly comparable

  • ProteinGym Hard

    Claude Opus 547.7%
    Source
    A.X series—

    Not directly comparable

  • Protein Design

    Claude Opus 542.5%
    Source
    A.X series—

    Not directly comparable

  • Organic chemistry V2

    Claude Opus 561.6%
    Source
    A.X series—

    Not directly comparable

  • Protocols (troubleshooting)

    Claude Opus 561.1%
    Source
    A.X series—

    Not directly comparable

  • Protocols (understanding)

    Claude Opus 578.4%
    Source
    A.X series—

    Not directly comparable

  • GPQA Diamond (Vals)

    Claude Opus 593.4%
    Source
    A.X series—

    Not directly comparable

  • MMLU-Pro (Vals)

    Claude Opus 591.6%
    Source
    A.X series—

    Not directly comparable

Math

  • IMO 2026

    Claude Opus 542/42
    Source
    A.X series—

    Not directly comparable

  • RiemannBench (no tools)

    Claude Opus 560.0%
    Source
    A.X series—

    Not directly comparable

  • RiemannBench (tools)

    Claude Opus 579.0%
    Source
    A.X series—

    Not directly comparable

  • ArXivMath Jun. 2026 (no tools)

    Claude Opus 590.8%
    Source
    A.X series—

    Not directly comparable

  • ArXivMath Jun. 2026 (tools)

    Claude Opus 591.3%
    Source
    A.X series—

    Not directly comparable

Multilingual

  • GMMLU

    Claude Opus 592.5%
    Source
    A.X series—

    Not directly comparable

  • MILU

    Claude Opus 592.1%
    Source
    A.X series—

    Not directly comparable

  • INCLUDE

    Claude Opus 589.8%
    Source
    A.X series—

    Not directly comparable

Multimodal

  • Chartography (no tools)

    Claude Opus 529.6%
    Source
    A.X series—

    Not directly comparable

  • Chartography (tools)

    Claude Opus 583.0%
    Source
    A.X series—

    Not directly comparable

  • BenchCAD Vision2Code (no tools)

    Claude Opus 50.366
    Source
    A.X series—

    Not directly comparable

  • BenchCAD Vision2Code (tools)

    Claude Opus 50.821
    Source
    A.X series—

    Not directly comparable

  • GDP.pdf (no tools)

    Claude Opus 583.4%
    Source
    A.X series—

    Not directly comparable

  • GDP.pdf (tools)

    Claude Opus 585.5%
    Source
    A.X series—

    Not directly comparable

  • OfficeQA

    Claude Opus 578.1%
    Source
    A.X series—

    Not directly comparable

  • OfficeQA Pro

    Claude Opus 566.9%
    Source
    A.X series—

    Not directly comparable

Frequently asked questions

Which is better, Claude Opus 5 or A.X series?

The public evidence has no benchmark result shared by both models, so it does not support a quality verdict. The page therefore keeps the decision tied to the specific documented workload.

Which is better for coding, Claude Opus 5 or A.X series?

A.X series is not ranked on the public lane for coding, so no winner is named for coding.

Which is better for agentic tasks, Claude Opus 5 or A.X series?

A.X series is not ranked on the public lane for agentic tasks, so no winner is named for agentic tasks.

Which costs less, Claude Opus 5 or A.X series?

Both models do not have comparable published API token rates, so this page does not name a universal price winner.

Which has the larger context window, Claude Opus 5 or A.X series?

Claude Opus 5 has the larger documented context window: 1M, compared with 64K.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated September 4, 2026

Watch Claude Opus 5 vs A.X series

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

Read a sample issue

Join 2,000+ readers.