LogoLMSpeed
Back to models

Model compare

Claude Opus 4.6 vs GPT-5.4 Mini

The readout for Claude Opus 4.6 and GPT-5.4 Mini, before the detailed comparison sheet.

Model A

Claude Opus 4.6

Anthropic

Leading
vs

Model B

GPT-5.4 Mini

OpenAI

Contender

Key Takeaways

Weighted outcome: Claude Opus 4.6. Benchmark capability categories carry 80%, while price, API performance, and availability carry 20%.

Decision read

Claude Opus 4.6

Claude Opus 4.6 has the higher weighted result; Model A / B score 53.7 to 46.3.

Evidence depth

131 data points

Includes 33 benchmark rows, 0 audit samples, and 9 provider examples.

Selection signal

Start with Claude Opus 4.6

The charts below split 42 high-signal samples across speed, scores, and audit health.

Change comparison

Switch either side of this report to compare another model with the same LMSpeed data pipeline.

Select a different model to open a new comparison URL.

Comparison sheet

This report only uses LMSpeed data for Claude Opus 4.6 and GPT-5.4 Mini: pricing, speed aggregates, third-party benchmark scores, and shared provider samples.

Model compare
Claude Opus 4.6
GPT-5.4 Mini
Overall leaderLeadingContender
Weighted overall score53.7 pts46.3 pts
Benchmark category leads4 categories3 categories
Operational advantagesCheapest input price, Free providersAverage speed, First-token latency, Provider coverage
Context window1M tokens400K tokens
Max output128K tokens128K tokens
Modalities

Input

TextImageFile

Output

Text

Input

FileImageText

Output

Text
Features
Text inputImage inputFile inputText outputTool callingStructured outputsJSON modeReasoning
Text inputImage inputFile inputText outputTool callingStructured outputsJSON modeReasoning

The overall result weights benchmark capability categories at 80% and price, API speed/latency, and availability at 20%. Recent test volume does not affect the winner, and missing benchmark categories are excluded.

Model metadata

Model compare
Claude Opus 4.6
GPT-5.4 Mini
DeveloperAnthropicOpenAI
ReleasedFeb 2026Mar 2026
ParametersNo dataNo data
TokenizerClaudeGPT
Knowledge cutoffNo data2025-08-31
OpenRouter IDanthropic/claude-opus-4.6openai/gpt-5.4-mini
ReferencesNo dataNo data

When to choose each model

This report only uses LMSpeed data for Claude Opus 4.6 and GPT-5.4 Mini: pricing, speed aggregates, third-party benchmark scores, and shared provider samples.

Claude Opus 4.6

Claude Opus 4.6 is stronger in benchmark categories (Agents, Knowledge, Math, Multimodal) and operational dimensions (Cheapest input price, Free providers).

GPT-5.4 Mini

GPT-5.4 Mini is stronger in benchmark categories (Coding, Reasoning, Instruction following) and operational dimensions (Average speed, First-token latency, Provider coverage).

Benchmark score comparison

Third-party benchmark profile synced into LMSpeed; only metrics available for both models are shown.

Category performance

Compare benchmark category scores on a 0-100 scale. Select a category to inspect the gap.

Model A coverage
8 / 8
Model B coverage
7 / 8
Shared
7 shared categories

Avg. score

Claude Opus 4.6

51.8

Avg. score

GPT-5.4 Mini

51.1

Agents

Claude Opus 4.6 leads by 4.2

Claude Opus 4.654.3
GPT-5.4 Mini50.1

Coding

GPT-5.4 Mini leads by 0.2

Claude Opus 4.656.6
GPT-5.4 Mini56.8

Reasoning

GPT-5.4 Mini leads by 2.0

Claude Opus 4.653.4
GPT-5.4 Mini55.4

Knowledge

Claude Opus 4.6 leads by 7.9

Claude Opus 4.653.3
GPT-5.4 Mini45.4

Math

Claude Opus 4.6 leads by 6.4

Claude Opus 4.655.7
GPT-5.4 Mini49.3

Multilingual

Claude Opus 4.6

Claude Opus 4.648.4
GPT-5.4 Mini-

Multimodal

Claude Opus 4.6 leads by 0.7

Claude Opus 4.647.8
GPT-5.4 Mini47.1

Instruction following

GPT-5.4 Mini leads by 9.0

Claude Opus 4.644.8
GPT-5.4 Mini53.8

Professional benchmark details

Metric-level scores with benchmark source, rank depth, confidence, error, and evaluation date where available.

Aggregatereported

BenchLM overall score

SourceClaude Opus 4.6

Claude Opus 4.6

winner

62.0

Rank #41/111 · confidence 4 · eval date 2026-02-01

+2.0

GPT-5.4 Mini

60.0

Rank #47/111 · confidence 4 · eval date 2026-03-17

Pricingverified

Input price

SourceClaude Opus 4.6 (Non-reasoning, High Effort)

Claude Opus 4.6

$5.00/M

Rank #168/186 · confidence 4

+$4.25/M

GPT-5.4 Mini

winner

$0.750/M

Rank #107/186 · confidence 4

Pricingverified

Blended price

SourceClaude Opus 4.6 (Non-reasoning, High Effort)

Claude Opus 4.6

$10.00/M

Rank #169/186 · confidence 4

+$8.31/M

GPT-5.4 Mini

winner

$1.69/M

Rank #113/186 · confidence 4

Pricingverified

Output price

SourceClaude Opus 4.6 (Non-reasoning, High Effort)

Claude Opus 4.6

$25.00/M

Rank #169/186 · confidence 4

+$20.50/M

GPT-5.4 Mini

winner

$4.50/M

Rank #125/186 · confidence 4

Reasoningreported

BenchLM Reasoning score

SourceClaude Opus 4.6

Claude Opus 4.6

67.1

Rank #42/67 · confidence 4 · eval date 2026-02-01

+6.8

GPT-5.4 Mini

winner

73.9

Rank #29/67 · confidence 4 · eval date 2026-03-17

Reasoningreported

CritPt

SourceClaude Opus 4.6

Claude Opus 4.6

2.8

Rank #57/88 · confidence 4 · eval date 2026-02-01

+7.2

GPT-5.4 Mini

winner

10.0

Rank #33/88 · confidence 4 · eval date 2026-03-17

Reasoningreported

AA-LCR

SourceClaude Opus 4.6

Claude Opus 4.6

67.0

Rank #71/106 · confidence 4 · eval date 2026-02-01

+10.0

GPT-5.4 Mini

winner

77.0

Rank #43/106 · confidence 4 · eval date 2026-03-17

Reasoningverified

GPQA

SourceClaude Opus 4.6 (Non-reasoning, High Effort)

Claude Opus 4.6

winner

84.0%

Rank #75/218 · confidence 4

+1.7%

GPT-5.4 Mini

82.3%

Rank #87/218 · confidence 4

Reasoningverified

HLE

SourceClaude Opus 4.6 (Non-reasoning, High Effort)

Claude Opus 4.6

winner

19.1%

Rank #89/216 · confidence 4

+0.5%

GPT-5.4 Mini

18.6%

Rank #91/216 · confidence 4

Codingreported

Vibe Code Bench

SourceClaude Opus 4.6

Claude Opus 4.6

winner

57.6

Rank #5/33 · confidence 4 · eval date 2026-02-01

+9.6

GPT-5.4 Mini

48.0

Rank #10/33 · confidence 4 · eval date 2026-03-17

Codingreported

FrontierCode 1.1 Main

SourceClaude Opus 4.6

Claude Opus 4.6

26.9

Rank #10/11 · confidence 4 · eval date 2026-02-01

+0.1

GPT-5.4 Mini

winner

27.0

Rank #9/11 · confidence 4 · eval date 2026-03-17

Codingreported

AA-SciCode

SourceClaude Opus 4.6

Claude Opus 4.6

45.7

Rank #54/113 · confidence 4 · eval date 2026-02-01

+6.4

GPT-5.4 Mini

winner

52.1

Rank #28/113 · confidence 4 · eval date 2026-03-17

Codingreported

BenchLM Coding score

SourceClaude Opus 4.6

Claude Opus 4.6

50.3

Rank #54/87 · confidence 4 · eval date 2026-02-01

+5.7

GPT-5.4 Mini

winner

56.0

Rank #36/87 · confidence 4 · eval date 2026-03-17

Mathreported

FrontierMath v2 (Tiers 1-3)

SourceClaude Opus 4.6

Claude Opus 4.6

winner

40.7

Rank #10/47 · confidence 4 · eval date 2026-02-01

+12.4

GPT-5.4 Mini

28.3

Rank #20/47 · confidence 4 · eval date 2026-03-17

Mathreported

FrontierMath v2 (Tier 4)

SourceClaude Opus 4.6

Claude Opus 4.6

winner

22.9

Rank #11/36 · confidence 4 · eval date 2026-02-01

+20.8

GPT-5.4 Mini

2.1

Rank #36/36 · confidence 4 · eval date 2026-03-17

Mathreported

BenchLM Math score

SourceClaude Opus 4.6

Claude Opus 4.6

winner

58.6

Rank #29/63 · confidence 4 · eval date 2026-02-01

+14.1

GPT-5.4 Mini

44.5

Rank #39/63 · confidence 4 · eval date 2026-03-17

Knowledgereported

HLE w/o tools

SourceClaude Opus 4.6

Claude Opus 4.6

winner

40.0

Rank #14/22 · confidence 4 · eval date 2026-02-01

+11.8

GPT-5.4 Mini

28.2

Rank #21/22 · confidence 4 · eval date 2026-03-17

Knowledgereported

BenchLM Knowledge score

SourceClaude Opus 4.6

Claude Opus 4.6

winner

79.8

Rank #15/83 · confidence 4 · eval date 2026-02-01

+22.8

GPT-5.4 Mini

57.0

Rank #62/83 · confidence 4 · eval date 2026-03-17

Knowledgereported

AA-Omniscience Hallucination Rate

SourceClaude Opus 4.6

Claude Opus 4.6

80.1

Rank #43/106 · confidence 4 · eval date 2026-02-01

+10.1

GPT-5.4 Mini

winner

90.2

Rank #18/106 · confidence 4 · eval date 2026-03-17

Knowledgereported

AA-Omniscience Accuracy

SourceClaude Opus 4.6

Claude Opus 4.6

winner

45.8

Rank #23/106 · confidence 4 · eval date 2026-02-01

+8.3

GPT-5.4 Mini

37.5

Rank #44/106 · confidence 4 · eval date 2026-03-17

Knowledgereported

AA-GPQA Diamond

SourceClaude Opus 4.6

Claude Opus 4.6

84.0

Rank #67/113 · confidence 4 · eval date 2026-02-01

+3.5

GPT-5.4 Mini

winner

87.5

Rank #46/113 · confidence 4 · eval date 2026-03-17

Knowledgereported

AA-HLE

SourceClaude Opus 4.6

Claude Opus 4.6

19.1

Rank #74/114 · confidence 4 · eval date 2026-02-01

+9.0

GPT-5.4 Mini

winner

28.1

Rank #54/114 · confidence 4 · eval date 2026-03-17

Knowledgereported

Artificial Analysis Intelligence Index

SourceClaude Opus 4.6

Claude Opus 4.6

winner

26.4

Rank #57/117 · confidence 4 · eval date 2026-02-01

+1.8

GPT-5.4 Mini

24.6

Rank #69/117 · confidence 4 · eval date 2026-03-17

Multimodalreported

MMMU-Pro

SourceClaude Opus 4.6

Claude Opus 4.6

winner

77.3

Rank #21/31 · confidence 4 · eval date 2026-02-01

+0.7

GPT-5.4 Mini

76.6

Rank #22/31 · confidence 4 · eval date 2026-03-17

Multimodalreported

BenchLM Multimodal Grounded score

SourceClaude Opus 4.6

Claude Opus 4.6

winner

59.5

Rank #40/56 · confidence 4 · eval date 2026-02-01

+2.3

GPT-5.4 Mini

57.2

Rank #41/56 · confidence 4 · eval date 2026-03-17

Multimodalreported

AA-MMMU-Pro

SourceClaude Opus 4.6

Claude Opus 4.6

72.5

Rank #45/68 · confidence 4 · eval date 2026-02-01

+0.8

GPT-5.4 Mini

winner

73.3

Rank #42/68 · confidence 4 · eval date 2026-03-17

Instruction followingreported

BenchLM Instruction Following score

SourceClaude Opus 4.6

Claude Opus 4.6

52.6

Rank #49/52 · confidence 4 · eval date 2026-02-01

+37.2

GPT-5.4 Mini

winner

89.8

Rank #17/52 · confidence 4 · eval date 2026-03-17

Instruction followingreported

AA-IFBench

SourceClaude Opus 4.6

Claude Opus 4.6

44.6

Rank #59/84 · confidence 4 · eval date 2026-02-01

+28.7

GPT-5.4 Mini

winner

73.3

Rank #23/84 · confidence 4 · eval date 2026-03-17

Agentsreported

APEX-Agents-AA

SourceClaude Opus 4.6 (Adaptive)

Claude Opus 4.6

winner

33.0

Rank #8/23 · confidence 1 · eval date 2026-02-01

+4.8

GPT-5.4 Mini

28.2

Rank #11/23 · confidence 4 · eval date 2026-03-17

Agentsreported

OSWorld-Verified

SourceClaude Opus 4.6

Claude Opus 4.6

winner

72.7

Rank #15/27 · confidence 4 · eval date 2026-02-01

+0.6

GPT-5.4 Mini

72.1

Rank #16/27 · confidence 4 · eval date 2026-03-17

Agentsreported

Terminal-Bench 2.0

SourceClaude Opus 4.6

Claude Opus 4.6

winner

65.4

Rank #24/53 · confidence 4 · eval date 2026-02-01

+5.4

GPT-5.4 Mini

60.0

Rank #31/53 · confidence 4 · eval date 2026-03-17

Agentsreported

τ²-bench results

SourceClaude Opus 4.6

Claude Opus 4.6

84.8

Rank #46/82 · confidence 4 · eval date 2026-02-01

+8.6

GPT-5.4 Mini

winner

93.4

Rank #28/82 · confidence 4 · eval date 2026-03-17

Agentsreported

BenchLM Agentic score

SourceClaude Opus 4.6

Claude Opus 4.6

winner

63.5

Rank #36/76 · confidence 4 · eval date 2026-02-01

+7.9

GPT-5.4 Mini

55.6

Rank #48/76 · confidence 4 · eval date 2026-03-17

API audit comparison

Latest completed audits from shared providers, with four safety and integrity score groups plus report links.

Provider
Claude Opus 4.6
GPT-5.4 Mini
No completed audits are available from shared providers yet.

Provider examples

Speed aggregates and input/output pricing share each provider row for real API selection and migration cost checks.

Provider
Claude Opus 4.6
GPT-5.4 Mini
XShuLab Sub2API25 tests

Claude Opus 4.6

speed / latency

N/A / N/A

input / output

No data

GPT-5.4 Mini

speed / latency

139 tok/s / 2844ms

input / output

No data

发现AI22 tests

Claude Opus 4.6

speed / latency

39 tok/s / 9397ms

input / output

No data

GPT-5.4 Mini

speed / latency

108 tok/s / 1950ms

input / output

No data

Kunkunout API20 tests

Claude Opus 4.6

speed / latency

40 tok/s / 16141ms

input / output

No data

GPT-5.4 Mini

speed / latency

N/A / N/A

input / output

No data

钠 API20 tests

Claude Opus 4.6

claude-opus-4-6

speed / latency

42 tok/s / 2282ms

input / output

$0.0010/M/$0.0051/M

GPT-5.4 Mini

gpt-5.4-mini

speed / latency

N/A / N/A

input / output

$0.011/M/$0.068/M

Rnglg2 API15 tests

Claude Opus 4.6

speed / latency

57 tok/s / 3926ms

input / output

No data

GPT-5.4 Mini

speed / latency

N/A / N/A

input / output

No data

Moyanjdc API

Claude Opus 4.6

claude-opus-4-6

speed / latency

No data

input / output

$0/M/$0/M

GPT-5.4 Mini

gpt-5.4-mini

speed / latency

No data

input / output

$0/M/$0/M

兔子API

Claude Opus 4.6

claude-opus-4-6

speed / latency

No data

input / output

$0/request

GPT-5.4 Mini

gpt-5.4-mini

speed / latency

No data

input / output

$0/request

Crond

Claude Opus 4.6

claude-opus-4-6-免费

speed / latency

No data

input / output

$0.0030/request

GPT-5.4 Mini

gpt-5.4-mini-免费

speed / latency

No data

input / output

$0.0030/request

BUZZ

Claude Opus 4.6

claude-opus-4-6

speed / latency

No data

input / output

$0.059/M/$0.294/M

GPT-5.4 Mini

gpt-5.4-mini

speed / latency

No data

input / output

$0.0044/M/$0.026/M

FAQ

Weighted outcome: Claude Opus 4.6. Benchmark capability categories carry 80%, while price, API performance, and availability carry 20%.

Why is this comparison indexable?
It has 6 verifiable comparison points, and both models have pricing or benchmark data.
Are missing metrics invented?
No. Metrics without LMSpeed data are omitted from this report.

Rankings are based on community-submitted tests and periodic health probes. Advisory only, not official data.