Enter opens the highlighted result. Esc closes.

Claude Opus 4.8 vs GPT-5.4

Claude Opus 4.8 leads GPT-5.4 on coding, 65.0 to 64.4 on the Vibe Coding Index, and GPT-5.4 costs 1.7x less per output token ($15.00 vs $25.00 per million). Updated October 22, 2026; 99 benchmarks have a result for both.

Replace Claude Opus 4.8 Replace GPT-5.4
Claude Opus 4.8Anthropic
Coding
65.0
Overall
69.8
Output $/M
$25.00
64
GPT-5.4OpenAI
Coding
64.4
Overall
69.6
Output $/M
$15.00
34
vs boards won of 99 shared, 1 tied

Claude Opus 4.8 or GPT-5.4: which to pick

Claude Opus 4.8 and GPT-5.4 are level for coding, 65.0 against 64.4 on the Vibe Coding Index, so price and context decide.

Pick GPT-5.4 for output tokens 1.7x cheaper ($15.00 vs $25.00 per million) and a lower cost per solved task ($1.02 against $2.17). Pick Claude Opus 4.8 for work measured by LMArena WebDev or AA Coding Index, where it leads.

Model AClaude Opus 4.8Anthropic65.0/1006 independent benchmarks$2.17 per solved taskModel BGPT-5.4OpenAI64.4/1003 independent benchmarks$1.02 per solved task
Shared results
99
Claude Opus 4.8 only
31
GPT-5.4 only
11
Like-for-like categories
8 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
Claude Opus 4.8 65.0 from 6 benchmarks, GPT-5.4 64.4 from 3 benchmarks.
Widest category
Agentic work, 14.8 points apart on average over 10 shared benchmarks.
Widest single gap
Design Arena: agents building full-stack apps: Claude Opus 4.8 1249 against GPT-5.4 1017, 29.0 points apart on a 0 to 100 scale.
Output price
Claude Opus 4.8 $25.00 against GPT-5.4 $15.00 per million output tokens, 1.7x apart.
Context window
Claude Opus 4.8 takes 1M tokens, GPT-5.4 1.05M.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryClaude Opus 4.8BarsGPT-5.4Ties
Coding132
Intelligence20
Agentic work82
Reasoning781 tie
Math51
Knowledge74
Web design80
General1417
All shared boards64341 tie
CodingLike-for-likeSix of these boards feed the Vibe Coding Index132
Coding benchmarks for Claude Opus 4.8 and GPT-5.4
BenchmarkClaude Opus 4.8GPT-5.4Leader
LiveBench coding77.3%Not measuredOnly Claude Opus 4.8 measured
LMArena WebDev15561464Claude Opus 4.8
AA Coding Index74.371.1Claude Opus 4.8
LiveCodeBench (Vals AI)87.8%84.1%Claude Opus 4.8
SWE-bench Verified (Vals AI)88.6%78.2%Claude Opus 4.8
Terminal-Bench 4 (Vals AI)23.2%Not measuredOnly Claude Opus 4.8 measured
Vibe Code Bench (Vals AI)82.7%67.4%Claude Opus 4.8
Code migration (Vals AI)47.3%35.0%Claude Opus 4.8
ProgramBench (Vals AI)1.0%0.5%Claude Opus 4.8
Terminal-Bench 2.1 (Vals AI)71.9%Not measuredOnly Claude Opus 4.8 measured
SkillsBench (Vals AI)59.2%51.7%Claude Opus 4.8
SciCode54.4%Not measuredOnly Claude Opus 4.8 measured
Terminal-Bench Hard (Artificial Analysis run)58.3%57.6%Claude Opus 4.8
Terminal-BenchNot measured81.8%Only GPT-5.4 measured
FrontierCode46.5%Not measuredOnly Claude Opus 4.8 measured
GSO-Bench47.1%31.4%Claude Opus 4.8
WeirdML82.9%77.7%Claude Opus 4.8
MirrorCodeNot measured15.6%Only GPT-5.4 measured
ALE-BenchNot measured1607.0Only GPT-5.4 measured
LMArena Text: coding14901496GPT-5.4
LMArena Text: software and IT14881491GPT-5.4
LMArena WebDev: React15571454Claude Opus 4.8
LMArena WebDev: HTML15431478Claude Opus 4.8
LMArena WebDev: from an imageNot measured1436Only GPT-5.4 measured
SciCode12.3%Not measuredOnly Claude Opus 4.8 measured
Neither model has a result on the other 14 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for Claude Opus 4.8 and GPT-5.4
BenchmarkClaude Opus 4.8GPT-5.4Leader
AA Intelligence Index41.839.0Claude Opus 4.8
Epoch Capabilities Index159.0157.3Claude Opus 4.8
Agentic workLike-for-likePublic boards, shown as published, kept out of the index82
Agentic work benchmarks for Claude Opus 4.8 and GPT-5.4
BenchmarkClaude Opus 4.8GPT-5.4Leader
AA Agentic Index41.9Not measuredOnly Claude Opus 4.8 measured
Design Arena: agents building web apps12331059Claude Opus 4.8
Design Arena: agents building full-stack apps12491017Claude Opus 4.8
Design Arena: agents building mobile apps12221087Claude Opus 4.8
Design Arena: agents building Android apps12331031Claude Opus 4.8
Design Arena: agents building Godot games12311135Claude Opus 4.8
Design Arena: agents building games1239Not measuredOnly Claude Opus 4.8 measured
Design Arena: HTML slides1171Not measuredOnly Claude Opus 4.8 measured
Design Arena: PowerPoint slides1202Not measuredOnly Claude Opus 4.8 measured
Design Arena: spreadsheets1303Not measuredOnly Claude Opus 4.8 measured
LiveBench research81.8%Not measuredOnly Claude Opus 4.8 measured
LiveBench browser96.0%Not measuredOnly Claude Opus 4.8 measured
LiveBench long agentic48.3%Not measuredOnly Claude Opus 4.8 measured
Finance Agent (Vals AI)53.9%Not measuredOnly Claude Opus 4.8 measured
Tax Agent Bench (Vals AI)30.0%Not measuredOnly Claude Opus 4.8 measured
Legal Research Bench (Vals AI)43.8%Not measuredOnly Claude Opus 4.8 measured
Excel Modeling Benchmark (Vals AI)69.4%Not measuredOnly Claude Opus 4.8 measured
Terminal-Bench-Science (Vals AI)4.3%Not measuredOnly Claude Opus 4.8 measured
Time Horizon Index (Vals AI)11.8%Not measuredOnly Claude Opus 4.8 measured
τ²-Bench Telecom94.4%87.1%Claude Opus 4.8
APEX-Agents48.9%52.4%GPT-5.4
METR time horizonsNot measured74.3%Only GPT-5.4 measured
Remote Labor Index8.3%Not measuredOnly Claude Opus 4.8 measured
OSWorld 2.020.6%Not measuredOnly Claude Opus 4.8 measured
DeepResearch Bench50.2%35.1%Claude Opus 4.8
PostTrainBench33.8%19.0%Claude Opus 4.8
Vending-Bench 2$5,787$6,144GPT-5.4
LMArena Search1204Not measuredOnly Claude Opus 4.8 measured
Neither model has a result on the other 13 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index78
Reasoning benchmarks for Claude Opus 4.8 and GPT-5.4
BenchmarkClaude Opus 4.8GPT-5.4Leader
GPQA Diamond92.4%91.7%Claude Opus 4.8
LiveBench reasoning89.2%Not measuredOnly Claude Opus 4.8 measured
MMMU Pro (Vals AI)86.6%87.5%GPT-5.4
GPQA Diamond (Artificial Analysis run)92.0%92.0%Tie
Humanity's Last Exam (Artificial Analysis run)48.7%43.7%Claude Opus 4.8
GPQA Diamond (Epoch AI run)91.0%93.3%GPT-5.4
ARC-AGI92.5%93.7%GPT-5.4
ARC-AGI-272.1%74.0%GPT-5.4
SimpleBench64.8%Not measuredOnly Claude Opus 4.8 measured
Chess puzzles52.0%48.0%Claude Opus 4.8
Mystery game puzzles36.0%37.0%GPT-5.4
EBR-bench28.6%25.4%Claude Opus 4.8
LMArena Text: hard prompts14821485GPT-5.4
LMArena Text: expert15031507GPT-5.4
LMArena Text: hard prompts in English14871479Claude Opus 4.8
LMArena Vision: diagrams13151327GPT-5.4
LMArena Vision: homework13371336Claude Opus 4.8
Kagi LLM Benchmark88.8%63.8%Claude Opus 4.8
Neither model has a result on the other 4 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index51
Math benchmarks for Claude Opus 4.8 and GPT-5.4
BenchmarkClaude Opus 4.8GPT-5.4Leader
AIMENot measured96.7%Only GPT-5.4 measured
OTIS Mock AIME98.3%97.8%Claude Opus 4.8
FrontierMath, tiers 1 to 380.0%78.6%Claude Opus 4.8
FrontierMath, tier 456.1%49.0%Claude Opus 4.8
ProofBench69.0%56.0%Claude Opus 4.8
LMArena Text: math14861488GPT-5.4
LMArena Text: mathematical work14971497Claude Opus 4.8
Neither model has a result on the other 6 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index74
Knowledge benchmarks for Claude Opus 4.8 and GPT-5.4
BenchmarkClaude Opus 4.8GPT-5.4Leader
MMLU Pro89.6%87.5%Claude Opus 4.8
LegalBench (Vals AI)83.6%86.0%GPT-5.4
CaseLaw (Vals AI)Not measured63.8%Only GPT-5.4 measured
CorpFin (Vals AI)66.7%65.3%Claude Opus 4.8
TaxEval (Vals AI)75.6%74.0%Claude Opus 4.8
MortgageTax (Vals AI)69.9%68.3%Claude Opus 4.8
MedQA (Vals AI)Not measured96.1%Only GPT-5.4 measured
MedCode (Vals AI)53.2%41.3%Claude Opus 4.8
SimpleQA Verified53.0%45.1%Claude Opus 4.8
LMArena Text: science14771479GPT-5.4
LMArena Text: medicine and healthcare14641458Claude Opus 4.8
LMArena Text: legal and government14771486GPT-5.4
LMArena Text: business and finance14581474GPT-5.4
FACTS Benchmark Suite53.5%Not measuredOnly Claude Opus 4.8 measured
Neither model has a result on the other 4 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index80
Web design benchmarks for Claude Opus 4.8 and GPT-5.4
BenchmarkClaude Opus 4.8GPT-5.4Leader
Design Arena: websites12581198Claude Opus 4.8
Design Arena: UI components12551212Claude Opus 4.8
Design Arena: game dev12641194Claude Opus 4.8
Design Arena: data viz12401196Claude Opus 4.8
Design Arena: 3D12221065Claude Opus 4.8
Design Arena: all code categories12551179Claude Opus 4.8
Design Arena: SVG11951141Claude Opus 4.8
Design Arena: ASCII art12771188Claude Opus 4.8
LiveBench design78.8%Not measuredOnly Claude Opus 4.8 measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index1417
General benchmarks for Claude Opus 4.8 and GPT-5.4
BenchmarkClaude Opus 4.8GPT-5.4Leader
LiveBench data analysis66.0%Not measuredOnly Claude Opus 4.8 measured
Public Benefits Bench (Vals AI)68.1%Not measuredOnly Claude Opus 4.8 measured
SAGE (Vals AI)54.8%43.3%Claude Opus 4.8
MedScribe (Vals AI)85.8%77.5%Claude Opus 4.8
Vals Multimodal Index70.9%Not measuredOnly Claude Opus 4.8 measured
IFBench62.2%73.9%GPT-5.4
Long context reasoning (AA-LCR)77.7%82.0%GPT-5.4
DTBench94.9%94.4%Claude Opus 4.8
LMCA57.5%52.0%Claude Opus 4.8
CL-benchNot measured27.9%Only GPT-5.4 measured
Surface Evolver Bench87.5%Not measuredOnly Claude Opus 4.8 measured
Furniture assembly42.5%37.5%Claude Opus 4.8
CL-bench LifeNot measured21.7%Only GPT-5.4 measured
LMArena Text14611469GPT-5.4
LMArena Text: instruction following14761469Claude Opus 4.8
LMArena Text: creative writing14541439Claude Opus 4.8
LMArena Vision12941303GPT-5.4
LMArena Text: English14681467Claude Opus 4.8
LMArena Text: non-English14501465GPT-5.4
LMArena Text: Chinese15061520GPT-5.4
LMArena Text: German14721472GPT-5.4
LMArena Text: Japanese14421485GPT-5.4
LMArena Text: Korean14311449GPT-5.4
LMArena Text: French14811493GPT-5.4
LMArena Text: Spanish14651452Claude Opus 4.8
LMArena Text: Russian14741480GPT-5.4
LMArena Text: Polish14591472GPT-5.4
LMArena Text: multi-turn14761482GPT-5.4
LMArena Text: longer queries14831473Claude Opus 4.8
LMArena Text: writing and language14611458Claude Opus 4.8
LMArena Text: entertainment and media14401438Claude Opus 4.8
LMArena Vision: English12931301GPT-5.4
LMArena Vision: Chinese13461354GPT-5.4
LMArena Vision: reading text13031312GPT-5.4
LMArena Vision: humor12821284GPT-5.4
LMArena Vision: recognizing thingsNot measured1251Only GPT-5.4 measured
LMArena Vision: creative writing13011283Claude Opus 4.8
LMArena Document14751471Claude Opus 4.8
Neither model has a result on the other 8 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Claude Opus 4.8 and GPT-5.4
BenchmarkClaude Opus 4.8GPT-5.4Leader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

Claude Opus 4.8$0.018Fits in one request
GPT-5.4$0.010Fits in one request

GPT-5.4 costs less

Repository review

50K input + 3K output tokens

Claude Opus 4.8$0.325Fits in one request
GPT-5.4$0.170Fits in one request

GPT-5.4 costs less

Agent loop

200K input + 10K output tokens

Claude Opus 4.8$1.25Fits in one request
GPT-5.4$0.650Fits in one request

GPT-5.4 costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of Claude Opus 4.8 and GPT-5.4
SpecificationAnthropic Claude Opus 4.8OpenAI GPT-5.4
MakerAnthropicOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedMay 27, 2026March 5, 2026
Context window1M1.05M
Input $/Mtok$5.00$2.50
Output $/Mtok$25.00$15.00
Coding index65.064.4
Independent measurements behind it63
Cost per solved task$2.17$1.02
SourceOpenRouter listingOpenRouter listing

Questions

Is Claude Opus 4.8 better than GPT-5.4 for coding?

Claude Opus 4.8 and GPT-5.4 are level for coding, 65.0 against 64.4 on the Vibe Coding Index, so price and context decide. Pick GPT-5.4 for output tokens 1.7x cheaper ($15.00 vs $25.00 per million) and a lower cost per solved task ($1.02 against $2.17). Pick Claude Opus 4.8 for work measured by LMArena WebDev or AA Coding Index, where it leads.

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

Claude Opus 4.8 costs $5.00 per million input tokens and $25.00 per million output tokens; GPT-5.4 costs $2.50 and $15.00. Per solved SWE-bench Verified task, Claude Opus 4.8 costs $2.17 and GPT-5.4 $1.02.

Which has the bigger context window, Claude Opus 4.8 or GPT-5.4?

GPT-5.4, with 1.05M tokens against 1M.

How many benchmarks is the Claude Opus 4.8 vs GPT-5.4 comparison based on?

99 benchmarks have a result for both models as of October 22, 2026. Claude Opus 4.8's coding index rests on 6 independent benchmarks and GPT-5.4's on 3. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/claude-opus-4-8-vs-gpt-5-4" title="Claude Opus 4.8 vs GPT-5.4, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/claude-opus-4-8-vs-gpt-5-4">LLMBenchmark.ai, Claude Opus 4.8 vs GPT-5.4</a></p>