Skip to main content
BenchLM
Advertise

Model comparison

Claude Opus 4.8 vs GPT-5.4

Data verified

Head-to-head evidence from 34 shared benchmark results across 7 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

78.34/100
Margin
4.1pts
← winning
GPT-5.4
OpenAI
74.24/100
5 category wins0 category wins

Public leaderboard positions: Claude Opus 4.8 #5 (Supported); GPT-5.4 #8 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Claude Opus 4.8 and GPT-5.4 share 34 comparable benchmark results. 5 of 8 categories are comparable. 19 results are unique to Claude Opus 4.8; 18 to GPT-5.4.

Updated July 23, 2026
Shared results
34
Claude Opus 4.8 only
19
GPT-5.4 only
18
Comparable categories
5 / 8

Pick Claude Opus 4.8 if you want the stronger benchmark profile. GPT-5.4 only becomes the better choice if you want the cheaper token bill or you need the larger 1.05M context window.

Confidence note. This is a partial-evidence comparison with 34 shared benchmark results across 7 evidence categories; 5 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

Claude Opus 4.8 is clearly ahead on the BenchAlign aggregate, 78.34 to 74.24. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Claude Opus 4.8's sharpest advantage is in coding, where it averages 81.1 against 57.7. The single biggest benchmark swing on the page is OfficeQA Pro, 66.2% to 53.2%.

Claude Opus 4.8 is also the more expensive model on tokens at $5.00 input / $25.00 output per 1M tokens, versus $2.50 input / $15.00 output per 1M tokens for GPT-5.4. GPT-5.4 gives you the larger context window at 1.05M, compared with 1M for Claude Opus 4.8.

Category breakdown

Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.

Category scores and score margins for Claude Opus 4.8 and GPT-5.4
CategoryClaude Opus 4.8ΔGPT-5.4
CodingClaude Opus 4.881.1Margin← 23.4GPT-5.457.7
MathClaude Opus 4.853.9Margin← 11.4GPT-5.442.5
KnowledgeClaude Opus 4.862.7Margin← 5.1GPT-5.457.6
MultimodalClaude Opus 4.877.0Margin← 3.8GPT-5.473.2
AgenticClaude Opus 4.880.3Margin← 3.1GPT-5.477.2

Decisive benchmark drivers

The largest measured benchmark gaps in this matchup, with exact reported values.

More
A · Claude Opus 4.8B · GPT-5.4

  1. OfficeQA Pro

    Multimodal
    Source ↗
    A 66.2%B 53.2%
    Winner: Claude Opus 4.8Δ 13
    OfficeQA Pro: Claude Opus 4.8 scored 66.2%; GPT-5.4 scored 53.2%. Claude Opus 4.8 wins this benchmark.

  2. SWE-bench Pro

    Coding
    Source ↗
    A 69.2%B 57.7%
    Winner: Claude Opus 4.8Δ 11.5
    SWE-bench Pro: Claude Opus 4.8 scored 69.2%; GPT-5.4 scored 57.7%. Claude Opus 4.8 wins this benchmark.

  3. OSWorld-Verified

    Agentic
    Source ↗
    A 83.4%B 75%
    Winner: Claude Opus 4.8Δ 8.4
    OSWorld-Verified: Claude Opus 4.8 scored 83.4%; GPT-5.4 scored 75%. Claude Opus 4.8 wins this benchmark.

  4. CharXiv

    Multimodal
    Source ↗
    A 89.9%B 82.8%
    Winner: Claude Opus 4.8Δ 7.1
    CharXiv: Claude Opus 4.8 scored 89.9%; GPT-5.4 scored 82.8%. Claude Opus 4.8 wins this benchmark.

  5. HLE

    Knowledge
    Source ↗
    A 57.9%B 52.1%
    Winner: Claude Opus 4.8Δ 5.8
    HLE: Claude Opus 4.8 scored 57.9%; GPT-5.4 scored 52.1%. Claude Opus 4.8 wins this benchmark.

Operational comparison

Runtime and commercial metrics are compared only when both models have a complete sourced value.

MetricClaude Opus 4.8GPT-5.4Comparison
Input / output priceUSD per 1M tokensClaude Opus 4.8$5 input / $25 outputGPT-5.4$2.5 input / $15 outputGPT-5.4 has the lower combined listed price.
Generation speedtokens per secondClaude Opus 4.8Not availableGPT-5.474 tok/sA complete speed comparison is not available.
First-answer latencyseconds to first tokenClaude Opus 4.8Not availableGPT-5.4151.79 sA complete latency comparison is not available.
Context windowmaximum listed tokensClaude Opus 4.81MGPT-5.41.05MGPT-5.4 lists the larger context window.

Benchmark Deep Dive

AgenticClaude Opus 4.8 wins
BenchmarkClaude Opus 4.8GPT-5.4Result
Terminal-Bench 2.0Source 74.6%75.1%GPT-5.4 leads
BrowseCompSource 84.3%82.7%Claude Opus 4.8 leads
DeepSearchQASource 93.1%73.6%Claude Opus 4.8 leads
OSWorld-VerifiedSource 83.4%75%Claude Opus 4.8 leads
Finance Agent v2Source 53.9%—Not comparable
GDPval-AASource 15941395Claude Opus 4.8 leads
MCP AtlasSource 82.2%70.6%Claude Opus 4.8 leads
ToolathlonSource 59.9%54.6%Claude Opus 4.8 leads
Gert LabsSource 72.97%64.89%Claude Opus 4.8 leads
AA Agentic IndexSource 47.2%41.1%Claude Opus 4.8 leads
τ²-bench resultsSource 94.4%87.1%Claude Opus 4.8 leads
GDPval-AASource 54.7%44.7%Claude Opus 4.8 leads
ResearchClawBenchSource 21.1%15.3%Claude Opus 4.8 leads
OSWorld 2.0Source 20.6%—Not comparable
AA BriefcaseSource 1347—Not comparable
AA AutomationBenchSource 48.5%—Not comparable
AA EnterpriseOps-GymSource 44.0%—Not comparable
AA Harvey LABSource 91.1%—Not comparable
AA Tau3 BankingSource 27.6%—Not comparable
terminalBenchHardSource 58.3%—Not comparable
aaTerminalBench21Source 84.6%—Not comparable
CyberGymSource —79.0%Not comparable
Claw-EvalSource —60.3%Not comparable
APEX-Agents-AASource —33.3%Not comparable
JobBenchSource —38.9%Not comparable
ExploitGymSource —6.0%Not comparable
CodingClaude Opus 4.8 wins
BenchmarkClaude Opus 4.8GPT-5.4Result
SWE-bench VerifiedSource 88.6%—Not comparable
SWE-bench ProSource 69.2%57.7%Claude Opus 4.8 leads
SWE MultilingualSource 84.4%—Not comparable
SWE MultimodalSource 38.4%—Not comparable
Terminal-Bench 2.0Source 74.6%—Not comparable
cursorBench31Source 58.4%—Not comparable
cursorBench32Source 62.3%—Not comparable
AA Coding IndexSource 74.3%71.0%Claude Opus 4.8 leads
AA-SciCodeSource 53.5%56.6%GPT-5.4 leads
FrontierCode 1.1 MainSource 46.5%—Not comparable
LiveCodeBench ProSource —87.5%Not comparable
React Native EvalsSource —85.3%Not comparable
Vibe Code BenchSource —67.42%Not comparable
Reasoning
BenchmarkClaude Opus 4.8GPT-5.4Result
AA-LCRSource 67.7%74.0%GPT-5.4 leads
CritPtSource 20.9%23.4%GPT-5.4 leads
KnowledgeClaude Opus 4.8 wins
BenchmarkClaude Opus 4.8GPT-5.4Result
GPQASource 93.6%92.8%Claude Opus 4.8 leads
GPQA-DSource 93.6%92.8%Claude Opus 4.8 leads
HLESource 57.9%52.1%Claude Opus 4.8 leads
HLE w/o toolsSource 49.8%39.8%Claude Opus 4.8 leads
Artificial Analysis Intelligence IndexSource 55.7%51.4%Claude Opus 4.8 leads
AA-GPQA DiamondSource 92.0%92.0%Tie
AA-HLESource 45.7%41.6%Claude Opus 4.8 leads
AA-Omniscience IndexSource 27.4%5.7%Claude Opus 4.8 leads
AA-Omniscience AccuracySource 46.6%50.0%GPT-5.4 leads
AA-Omniscience Hallucination RateSource 35.9%88.6%Claude Opus 4.8 leads
HealthBench HardSource —40.1%Not comparable
MedXpertQA (Text)Source —59.6%Not comparable
HealthBench ProfessionalSource —48.1%Not comparable
MathClaude Opus 4.8 wins
BenchmarkClaude Opus 4.8GPT-5.4Result
USAMO 2026Source 96.7%—Not comparable
FrontierMath v2 (Tiers 1-3)Source 47.241%47.600%GPT-5.4 leads
FrontierMath v2 (Tier 4)Source 31.250%27.100%Claude Opus 4.8 leads
Multilingual
BenchmarkClaude Opus 4.8GPT-5.4Result
INCLUDESource 87.6%—Not comparable
MultimodalClaude Opus 4.8 wins
BenchmarkClaude Opus 4.8GPT-5.4Result
OfficeQA ProSource 66.2%53.2%Claude Opus 4.8 leads
ScreenSpot ProSource 87.9%85.4%Claude Opus 4.8 leads
CharXivSource 89.9%82.8%Claude Opus 4.8 leads
CharXiv w/o toolsSource 80.5%—Not comparable
Design Arena WebsiteSource 12701250Claude Opus 4.8 leads
MMMU-ProSource —81.2%Not comparable
MMMU-Pro w/ PythonSource —82.1%Not comparable
ERQASource —65.4%Not comparable
SimpleVQASource —61.1%Not comparable
ZeroBenchSource —41.0%Not comparable
MedXpertQA (MM)Source —77.1%Not comparable
AA-MMMU-ProSource —78.4%Not comparable
Inst. Following
BenchmarkClaude Opus 4.8GPT-5.4Result
AA-IFBenchSource 62.2%73.9%GPT-5.4 leads
Frequently Asked Questions (6)

Which is better, Claude Opus 4.8 or GPT-5.4?

Claude Opus 4.8 is ahead on BenchLM's BenchAlign leaderboard, 78.34 to 74.24. The biggest single separator in this matchup is OfficeQA Pro, where the scores are 66.2% and 53.2%.

Which is better for knowledge tasks, Claude Opus 4.8 or GPT-5.4?

Claude Opus 4.8 has the edge for knowledge tasks in this comparison, averaging 62.7 versus 57.6. Inside this category, AA-Omniscience Hallucination Rate is the benchmark that creates the most daylight between them.

Which is better for coding, Claude Opus 4.8 or GPT-5.4?

Claude Opus 4.8 has the edge for coding in this comparison, averaging 81.1 versus 57.7. Inside this category, SWE-bench Pro is the benchmark that creates the most daylight between them.

Which is better for math, Claude Opus 4.8 or GPT-5.4?

Claude Opus 4.8 has the edge for math in this comparison, averaging 53.9 versus 42.5. Inside this category, FrontierMath v2 (Tier 4) is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Claude Opus 4.8 or GPT-5.4?

Claude Opus 4.8 has the edge for agentic tasks in this comparison, averaging 80.3 versus 77.2. Inside this category, GDPval-AA is the benchmark that creates the most daylight between them.

Which is better for multimodal and grounded tasks, Claude Opus 4.8 or GPT-5.4?

Claude Opus 4.8 has the edge for multimodal and grounded tasks in this comparison, averaging 77 versus 73.2. Inside this category, Design Arena Website is the benchmark that creates the most daylight between them.

Related Comparisons

Explore More

Anthropic Compare Pricing Methodology Find Your Best LLM Overall Rankings
Last updated: July 23, 2026

Choose a model with this week’s evidence

Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.

One email each week. Unsubscribe anytime.