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 4.6

Anthropic

67.98/100

Supported · Public rank #30

90% interval 53.4–82.5

Claude Opus 4.6 vs GPT-5.4

Updated September 3, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.

OpenAI logo
Model B
GPT-5.4

OpenAI

72.89/100

Supported · Public rank #20

90% interval 69.1–76.7

Decision reading

GPT-5.4 has the higher public score estimate, 72.89 versus 67.98, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.

25 results are shared. Category rows based on different benchmark sets are marked directional and do not name a winner.

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.

  • Agentic work

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

    GPT-5.4

    GPT-5.4 leads on the same 3 weighted benchmark rows.

    Confidence: stronger

  • Long documents

    Prompts that approach the documented context limit

    GPT-5.4

    GPT-5.4 has the larger documented context window.

    Confidence: documented

  • Chat turn cost

    1K fresh input + 500 output tokens

    GPT-5.4

    GPT-5.4 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

    GPT-5.4

    GPT-5.4 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

    GPT-5.4

    GPT-5.4 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

    The category averages use different weighted benchmark sets, so they are directional rather than like-for-like.

    Confidence: limited

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
25
Claude Opus 4.6 only
8
GPT-5.4 only
12
Like-for-like categories
2 / 8

3 categories use different evidence sets. Those rows remain visible for coverage context but do not name a winner.

Category results, on a stated basis

Each row states whether both averages use the same weighted benchmark set. Directional and not-comparable rows remain visible, but they never receive a winner in this template.

Agentic

Like-for-like
Claude Opus 4.6
73.0
GPT-5.4
77.2
Weighted basis
3 vs 3 rows
Reading
GPT-5.4 leads

Math

Like-for-like
Claude Opus 4.6
36.3
GPT-5.4
42.5
Weighted basis
2 vs 2 rows
Reading
GPT-5.4 leads

Coding

Directional only
Claude Opus 4.6
68.1
GPT-5.4
57.7
Weighted basis
3 vs 1 rows
Reading
Directional only

Knowledge

Directional only
Claude Opus 4.6
69.1
GPT-5.4
57.6
Weighted basis
4 vs 2 rows
Reading
Directional only

Multimodal

Directional only
Claude Opus 4.6
77.3
GPT-5.4
73.2
Weighted basis
1 vs 3 rows
Reading
Directional only

Reasoning

Not comparable
Claude Opus 4.6
Not measured
GPT-5.4
74.0
Weighted basis
0 vs 1 rows
Reading
Not comparable

Multilingual

Not comparable
Claude Opus 4.6
Not measured
GPT-5.4
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Instruction following

Not comparable
Claude Opus 4.6
Not measured
GPT-5.4
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

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.

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 4.6
$0.0175
Fits in one request
GPT-5.4
$0.01
Fits in one request

GPT-5.4 has the lower modeled cost

Costs use the listed standard API rates.

Repository review

50K fresh input + 3K output tokens

Claude Opus 4.6
$0.325
Fits in one request
GPT-5.4
$0.17
Fits in one request

GPT-5.4 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 Opus 4.6
$1.35
Fits in one request
Cached input priced at the published list-input rate
GPT-5.4
$0.25
Fits in one request

GPT-5.4 has the lower modeled cost

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

Claude Opus 4.6

1M

GPT-5.4

1.05M

OpenAI pricing

Cached-input rate

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

Claude Opus 4.6

Not published

GPT-5.4

$0.25 per 1M cached input tokens

OpenAI pricing

Documented inputs

Claude Opus 4.6

Not sourced

GPT-5.4

Not sourced

Documented outputs

Claude Opus 4.6

Not sourced

GPT-5.4

Not sourced

Provider availability

Claude Opus 4.6

Not sourced

GPT-5.4

Not sourced

Reasoning profile

Claude Opus 4.6

Non-Reasoning

GPT-5.4

Reasoning

Weight access

Claude Opus 4.6

Proprietary

GPT-5.4

Proprietary

License

Claude Opus 4.6

Proprietary

GPT-5.4

Proprietary

Release date

Claude Opus 4.6

2026-02-01

GPT-5.4

2026-03-05

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
GPT-5.4 has the higher public score estimate, 72.89 versus 67.98, but the 90% score intervals overlap.
Workload cost
Repository review: $0.325 vs $0.17. Cache-heavy agent loop: $1.35 vs $0.25.
Context tradeoff
GPT-5.4 has the larger documented window (1.05M).

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 evidence45 rows

Agentic

  • Terminal-Bench 2.0

    Claude Opus 4.665.4%
    Source
    GPT-5.475.1%
    Source

    GPT-5.4 leads this result

  • BrowseComp

    Claude Opus 4.683.7%
    Source
    GPT-5.482.7%
    Source

    Claude Opus 4.6 leads this result

  • OSWorld-Verified

    Claude Opus 4.672.7%
    Source
    GPT-5.475%
    Source

    GPT-5.4 leads this result

  • Claw-Eval

    Shared source
    Claude Opus 4.670.4%
    GPT-5.460.3%

    Claude Opus 4.6 leads this result

  • DeepSearchQA

    Shared source
    Claude Opus 4.673.7%
    GPT-5.473.6%

    Claude Opus 4.6 leads this result

  • CyberGym

    Claude Opus 4.666.6%
    Source
    GPT-5.479.0%
    Source

    GPT-5.4 leads this result

  • Gert Labs

    Shared source
    Claude Opus 4.661.85%
    GPT-5.464.89%

    GPT-5.4 leads this result

  • ResearchClawBench

    Shared source
    Claude Opus 4.619.9%
    GPT-5.415.3%

    Claude Opus 4.6 leads this result

  • JobBench

    Shared source
    Claude Opus 4.636.7%
    GPT-5.438.9%

    GPT-5.4 leads this result

  • MCP Atlas

    Claude Opus 4.6—
    GPT-5.470.6%
    Source

    Not directly comparable

  • Toolathlon

    Claude Opus 4.6—
    GPT-5.454.6%
    Source

    Not directly comparable

  • τ²-bench results

    Claude Opus 4.6—
    GPT-5.498.9%
    Source

    Not directly comparable

  • ExploitGym

    Claude Opus 4.6—
    GPT-5.46.0%
    Source

    Not directly comparable

Coding

  • SWE-bench Verified

    Claude Opus 4.680.8%
    Source
    GPT-5.4—

    Not directly comparable

  • SWE-bench Verified*

    Claude Opus 4.675.6%
    Source
    GPT-5.4—

    Not directly comparable

  • LiveCodeBench Pro

    Shared source
    Claude Opus 4.670.7%
    GPT-5.487.5%

    GPT-5.4 leads this result

  • SWE-bench Pro

    Claude Opus 4.653.4%
    Source
    GPT-5.457.7%
    Source

    GPT-5.4 leads this result

  • SWE-Rebench

    Claude Opus 4.665.3%
    Source
    GPT-5.4—

    Not directly comparable

  • React Native Evals

    Shared source
    Claude Opus 4.684.1%
    GPT-5.485.3%

    GPT-5.4 leads this result

  • Vibe Code Bench

    Shared source
    Claude Opus 4.657.57%
    GPT-5.467.42%

    GPT-5.4 leads this result

  • FrontierCode 1.1 Main

    Claude Opus 4.626.9%
    Source
    GPT-5.4—

    Not directly comparable

Reasoning

  • ARC-AGI-2

    Claude Opus 4.6—
    GPT-5.474.0%
    Source

    Not directly comparable

  • ARC-AGI-3

    Claude Opus 4.6—
    GPT-5.40.2%
    Source

    Not directly comparable

Knowledge

  • GPQA

    Claude Opus 4.691.3%
    Source
    GPT-5.492.8%
    Source

    GPT-5.4 leads this result

  • GPQA-D

    Claude Opus 4.689.2%
    Source
    GPT-5.492.8%
    Source

    GPT-5.4 leads this result

  • SuperGPQA

    Claude Opus 4.695%
    Source
    GPT-5.4—

    Not directly comparable

  • MMLU-Pro

    Claude Opus 4.682%
    Source
    GPT-5.4—

    Not directly comparable

  • MMLU-Pro (Arcee)

    Claude Opus 4.689.1%
    Source
    GPT-5.4—

    Not directly comparable

  • HLE

    Claude Opus 4.653%
    Source
    GPT-5.452.1%
    Source

    Claude Opus 4.6 leads this result

  • HLE w/o tools

    Claude Opus 4.640%
    Source
    GPT-5.439.8%
    Source

    Claude Opus 4.6 leads this result

  • HealthBench Hard

    Shared source
    Claude Opus 4.614.8%
    GPT-5.440.1%

    GPT-5.4 leads this result

  • MedXpertQA (Text)

    Shared source
    Claude Opus 4.652.1%
    GPT-5.459.6%

    GPT-5.4 leads this result

  • HealthBench Professional

    Claude Opus 4.6—
    GPT-5.448.1%
    Source

    Not directly comparable

Math

  • AIME25 (Arcee)

    Claude Opus 4.699.8%
    Source
    GPT-5.4—

    Not directly comparable

  • FrontierMath v2 (Tiers 1-3)

    Shared source
    Claude Opus 4.640.700%
    GPT-5.447.600%

    GPT-5.4 leads this result

  • FrontierMath v2 (Tier 4)

    Shared source
    Claude Opus 4.622.900%
    GPT-5.427.100%

    GPT-5.4 leads this result

Multimodal

  • MMMU-Pro

    Claude Opus 4.677.3%
    Source
    GPT-5.481.2%
    Source

    GPT-5.4 leads this result

  • ERQA

    Shared source
    Claude Opus 4.651.6%
    GPT-5.465.4%

    GPT-5.4 leads this result

  • ScreenSpot Pro

    Shared source
    Claude Opus 4.683.1%
    GPT-5.485.4%

    GPT-5.4 leads this result

  • MedXpertQA (MM)

    Shared source
    Claude Opus 4.664.8%
    GPT-5.477.1%

    GPT-5.4 leads this result

  • OfficeQA Pro

    Claude Opus 4.6—
    GPT-5.453.2%
    Source

    Not directly comparable

  • MMMU-Pro w/ Python

    Claude Opus 4.6—
    GPT-5.482.1%
    Source

    Not directly comparable

  • CharXiv

    Claude Opus 4.6—
    GPT-5.482.8%
    Source

    Not directly comparable

  • SimpleVQA

    Claude Opus 4.6—
    GPT-5.461.1%
    Source

    Not directly comparable

  • ZeroBench

    Claude Opus 4.6—
    GPT-5.441.0%
    Source

    Not directly comparable

Frequently asked questions

Which is better, Claude Opus 4.6 or GPT-5.4?

GPT-5.4 has the higher public score estimate, 72.89 versus 67.98, 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 Opus 4.6 or GPT-5.4?

The current coding averages use different weighted benchmark sets, so BenchLM does not name a winner from them. Read the shared benchmark rows directly and test the models on the same task set.

Which is better for agentic tasks, Claude Opus 4.6 or GPT-5.4?

GPT-5.4 leads the like-for-like agentic tasks comparison across 3 shared weighted benchmark rows.

Which costs less, Claude Opus 4.6 or GPT-5.4?

For the stated presets, chat costs $0.0175 on Claude Opus 4.6 and $0.01 on GPT-5.4; repository review costs $0.325 and $0.17; the cache-heavy agent loop costs $1.35 and $0.25. 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 Opus 4.6 or GPT-5.4?

GPT-5.4 has the larger documented context window: 1.05M, compared with 1M.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated September 3, 2026

Watch Claude Opus 4.6 vs GPT-5.4

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

Read a sample issue

Join 2,000+ readers.