Enter opens the highlighted result. Esc closes.

Claude Opus 4.8 vs GPT-5.5

GPT-5.5 leads Claude Opus 4.8 on coding, 71.3 to 65.0 on the Vibe Coding Index, and Claude Opus 4.8 costs 1.2x less per output token ($25.00 vs $30.00 per million). Updated October 22, 2026; 125 benchmarks have a result for both.

Replace Claude Opus 4.8 Replace GPT-5.5
Claude Opus 4.8Anthropic
Coding
65.0
Overall
69.8
Output $/M
$25.00
52
GPT-5.5OpenAI
Coding
71.3
Overall
69.2
Output $/M
$30.00
72
vs boards won of 125 shared, 1 tied

Claude Opus 4.8 or GPT-5.5: which to pick

Pick GPT-5.5 for coding: it scores 71.3 on the Vibe Coding Index against 65.0 for Claude Opus 4.8, though on 5 benchmarks against 6.

Pick Claude Opus 4.8 for output tokens 1.2x cheaper ($25.00 vs $30.00 per million) and work measured by LMArena WebDev or LiveCodeBench (Vals AI), where it leads.

Model AClaude Opus 4.8Anthropic65.0/1006 independent benchmarks$2.17 per solved taskModel BGPT-5.5OpenAI71.3/1005 independent benchmarks$1.65 per solved task
Shared results
125
Claude Opus 4.8 only
5
GPT-5.5 only
9
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-5.5 71.3 from 5 benchmarks.
Widest category
Agentic work, 10.0 points apart on average over 24 shared benchmarks.
Widest single gap
LiveBench browser: Claude Opus 4.8 96.0% against GPT-5.5 48.0%, 48.0 points apart on a 0 to 100 scale.
Output price
Claude Opus 4.8 $25.00 against GPT-5.5 $30.00 per million output tokens, 1.2x apart.
Context window
Claude Opus 4.8 takes 1M tokens, GPT-5.5 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-5.5Ties
Coding910
Intelligence11
Agentic work213
Reasoning4131 tie
Math33
Knowledge48
Web design45
General629
All shared boards52721 tie
CodingLike-for-likeSix of these boards feed the Vibe Coding Index910
Coding benchmarks for Claude Opus 4.8 and GPT-5.5
BenchmarkClaude Opus 4.8GPT-5.5Leader
LiveBench coding77.3%85.1%GPT-5.5
LMArena WebDev15561513Claude Opus 4.8
AA Coding Index74.374.9GPT-5.5
LiveCodeBench (Vals AI)87.8%85.3%Claude Opus 4.8
SWE-bench Verified (Vals AI)88.6%82.6%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%69.8%Claude Opus 4.8
Code migration (Vals AI)47.3%45.2%Claude Opus 4.8
ProgramBench (Vals AI)1.0%0.5%Claude Opus 4.8
Terminal-Bench 2.1 (Vals AI)71.9%76.4%GPT-5.5
SkillsBench (Vals AI)59.2%62.5%GPT-5.5
SciCode54.4%56.1%GPT-5.5
Terminal-Bench Hard (Artificial Analysis run)58.3%60.6%GPT-5.5
Terminal-BenchNot measured84.7%Only GPT-5.5 measured
FrontierCode46.5%43.0%Claude Opus 4.8
GSO-Bench47.1%40.2%Claude Opus 4.8
WeirdML82.9%84.9%GPT-5.5
MirrorCodeNot measured10.0%Only GPT-5.5 measured
ALE-BenchNot measured1943.0Only GPT-5.5 measured
LMArena Text: coding14901494GPT-5.5
LMArena Text: software and IT14881493GPT-5.5
LMArena WebDev: React15571506Claude Opus 4.8
LMArena WebDev: HTML15431543GPT-5.5
LMArena WebDev: from an imageNot measured1532Only GPT-5.5 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 index11
Intelligence benchmarks for Claude Opus 4.8 and GPT-5.5
BenchmarkClaude Opus 4.8GPT-5.5Leader
AA Intelligence Index41.838.4Claude Opus 4.8
Epoch Capabilities Index159.0159.7GPT-5.5
Agentic workLike-for-likePublic boards, shown as published, kept out of the index213
Agentic work benchmarks for Claude Opus 4.8 and GPT-5.5
BenchmarkClaude Opus 4.8GPT-5.5Leader
AA Agentic Index41.936.4Claude Opus 4.8
Design Arena: agents building web apps12331117Claude Opus 4.8
Design Arena: agents building full-stack apps12491089Claude Opus 4.8
Design Arena: agents building mobile apps12221142Claude Opus 4.8
Design Arena: agents building Android apps12331164Claude Opus 4.8
Design Arena: agents building Godot games12311173Claude Opus 4.8
Design Arena: agents building games12391179Claude Opus 4.8
Design Arena: HTML slides11711017Claude Opus 4.8
Design Arena: PowerPoint slides12021084Claude Opus 4.8
Design Arena: spreadsheets1303Not measuredOnly Claude Opus 4.8 measured
LiveBench research81.8%61.6%Claude Opus 4.8
LiveBench browser96.0%48.0%Claude Opus 4.8
LiveBench long agentic48.3%44.6%Claude Opus 4.8
SREBench (Vals AI)Not measured3.8%Only GPT-5.5 measured
Finance Agent (Vals AI)53.9%51.8%Claude Opus 4.8
Tax Agent Bench (Vals AI)30.0%21.8%Claude Opus 4.8
Legal Research Bench (Vals AI)43.8%40.4%Claude Opus 4.8
Excel Modeling Benchmark (Vals AI)69.4%64.5%Claude Opus 4.8
Terminal-Bench-Science (Vals AI)4.3%Not measuredOnly Claude Opus 4.8 measured
Time Horizon Index (Vals AI)11.8%8.3%Claude Opus 4.8
τ²-Bench Telecom94.4%93.9%Claude Opus 4.8
APEX-Agents48.9%55.1%GPT-5.5
Remote Labor Index8.3%6.3%Claude Opus 4.8
OSWorld 2.020.6%13.0%Claude Opus 4.8
DeepResearch Bench50.2%54.0%GPT-5.5
PostTrainBench33.8%27.2%Claude Opus 4.8
Vending-Bench 2$5,787$7,524GPT-5.5
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 index413
Reasoning benchmarks for Claude Opus 4.8 and GPT-5.5
BenchmarkClaude Opus 4.8GPT-5.5Leader
GPQA Diamond92.4%93.2%GPT-5.5
LiveBench reasoning89.2%89.7%GPT-5.5
MMMU Pro (Vals AI)86.6%88.3%GPT-5.5
GPQA Diamond (Artificial Analysis run)92.0%93.5%GPT-5.5
Humanity's Last Exam (Artificial Analysis run)48.7%45.8%Claude Opus 4.8
GPQA Diamond (Epoch AI run)91.0%90.7%Claude Opus 4.8
ARC-AGI92.5%95.0%GPT-5.5
ARC-AGI-272.1%85.0%GPT-5.5
SimpleBench64.8%69.0%GPT-5.5
Chess puzzles52.0%63.0%GPT-5.5
Mystery game puzzles36.0%56.0%GPT-5.5
EBR-bench28.6%34.3%GPT-5.5
LMArena Text: hard prompts14821489GPT-5.5
LMArena Text: expert15031507GPT-5.5
LMArena Text: hard prompts in English14871483Claude Opus 4.8
LMArena Vision: diagrams13151325GPT-5.5
LMArena Vision: homework13371332Claude Opus 4.8
Kagi LLM Benchmark88.8%88.8%Tie
Neither model has a result on the other 4 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index33
Math benchmarks for Claude Opus 4.8 and GPT-5.5
BenchmarkClaude Opus 4.8GPT-5.5Leader
OTIS Mock AIME98.3%84.4%Claude Opus 4.8
FrontierMath, tiers 1 to 380.0%85.3%GPT-5.5
FrontierMath, tier 456.1%72.5%GPT-5.5
ProofBench69.0%50.0%Claude Opus 4.8
LMArena Text: math14861487GPT-5.5
LMArena Text: mathematical work14971496Claude Opus 4.8
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index48
Knowledge benchmarks for Claude Opus 4.8 and GPT-5.5
BenchmarkClaude Opus 4.8GPT-5.5Leader
MMLU Pro89.6%88.1%Claude Opus 4.8
LegalBench (Vals AI)83.6%86.5%GPT-5.5
CaseLaw (Vals AI)Not measured66.2%Only GPT-5.5 measured
CorpFin (Vals AI)66.7%68.4%GPT-5.5
TaxEval (Vals AI)75.6%75.0%Claude Opus 4.8
MortgageTax (Vals AI)69.9%68.8%Claude Opus 4.8
MedCode (Vals AI)53.2%49.1%Claude Opus 4.8
SimpleQA Verified53.0%63.0%GPT-5.5
LMArena Text: science14771485GPT-5.5
LMArena Text: medicine and healthcare14641465GPT-5.5
LMArena Text: legal and government14771491GPT-5.5
LMArena Text: business and finance14581478GPT-5.5
FACTS Benchmark Suite53.5%66.7%GPT-5.5
Neither model has a result on the other 5 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index45
Web design benchmarks for Claude Opus 4.8 and GPT-5.5
BenchmarkClaude Opus 4.8GPT-5.5Leader
Design Arena: websites12581256Claude Opus 4.8
Design Arena: UI components12551257GPT-5.5
Design Arena: game dev12641293GPT-5.5
Design Arena: data viz12401264GPT-5.5
Design Arena: 3D12221197Claude Opus 4.8
Design Arena: all code categories12551260GPT-5.5
Design Arena: SVG11951234GPT-5.5
Design Arena: ASCII art12771255Claude Opus 4.8
LiveBench design78.8%70.3%Claude Opus 4.8
GeneralLike-for-likePublic boards, shown as published, kept out of the index629
General benchmarks for Claude Opus 4.8 and GPT-5.5
BenchmarkClaude Opus 4.8GPT-5.5Leader
LiveBench data analysis66.0%81.6%GPT-5.5
Public Benefits Bench (Vals AI)68.1%60.9%Claude Opus 4.8
SAGE (Vals AI)54.8%51.5%Claude Opus 4.8
MedScribe (Vals AI)85.8%86.9%GPT-5.5
Vals Multimodal Index70.9%68.1%Claude Opus 4.8
IFBench62.2%75.9%GPT-5.5
Long context reasoning (AA-LCR)77.7%84.3%GPT-5.5
DTBench94.9%96.0%GPT-5.5
LMCA57.5%54.3%Claude Opus 4.8
Surface Evolver Bench87.5%88.1%GPT-5.5
Furniture assembly42.5%44.2%GPT-5.5
ExploitBenchNot measured47.4%Only GPT-5.5 measured
CL-bench LifeNot measured22.2%Only GPT-5.5 measured
LMArena Text14611471GPT-5.5
LMArena Text: instruction following14761478GPT-5.5
LMArena Text: creative writing14541455GPT-5.5
LMArena Vision12941297GPT-5.5
LMArena Text: English14681470GPT-5.5
LMArena Text: non-English14501467GPT-5.5
LMArena Text: Chinese15061534GPT-5.5
LMArena Text: German14721480GPT-5.5
LMArena Text: Japanese14421499GPT-5.5
LMArena Text: Korean14311460GPT-5.5
LMArena Text: French14811485GPT-5.5
LMArena Text: Spanish14651468GPT-5.5
LMArena Text: Russian14741473Claude Opus 4.8
LMArena Text: Polish14591465GPT-5.5
LMArena Text: multi-turn14761476GPT-5.5
LMArena Text: longer queries14831484GPT-5.5
LMArena Text: writing and language14611466GPT-5.5
LMArena Text: entertainment and media14401449GPT-5.5
LMArena Vision: English12931292Claude Opus 4.8
LMArena Vision: Chinese13461360GPT-5.5
LMArena Vision: reading text13031306GPT-5.5
LMArena Vision: humor12821293GPT-5.5
LMArena Vision: recognizing thingsNot measured1231Only GPT-5.5 measured
LMArena Vision: creative writing13011301GPT-5.5
LMArena Document14751486GPT-5.5
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-5.5
BenchmarkClaude Opus 4.8GPT-5.5Leader
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-5.5$0.020Fits in one request

Claude Opus 4.8 costs less

Repository review

50K input + 3K output tokens

Claude Opus 4.8$0.325Fits in one request
GPT-5.5$0.340Fits in one request

Claude Opus 4.8 costs less

Agent loop

200K input + 10K output tokens

Claude Opus 4.8$1.25Fits in one request
GPT-5.5$1.30Fits in one request

Claude Opus 4.8 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-5.5
SpecificationAnthropic Claude Opus 4.8OpenAI GPT-5.5
MakerAnthropicOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedMay 27, 2026April 24, 2026
Context window1M1.05M
Input $/Mtok$5.00$5.00
Output $/Mtok$25.00$30.00
Coding index65.071.3
Independent measurements behind it65
Cost per solved task$2.17$1.65
SourceOpenRouter listingOpenRouter listing

Questions

Is Claude Opus 4.8 better than GPT-5.5 for coding?

Pick GPT-5.5 for coding: it scores 71.3 on the Vibe Coding Index against 65.0 for Claude Opus 4.8, though on 5 benchmarks against 6. Pick Claude Opus 4.8 for output tokens 1.2x cheaper ($25.00 vs $30.00 per million) and work measured by LMArena WebDev or LiveCodeBench (Vals AI), where it leads.

Which is cheaper, Claude Opus 4.8 or GPT-5.5?

Claude Opus 4.8 costs $5.00 per million input tokens and $25.00 per million output tokens; GPT-5.5 costs $5.00 and $30.00. Per solved SWE-bench Verified task, Claude Opus 4.8 costs $2.17 and GPT-5.5 $1.65.

Which has the bigger context window, Claude Opus 4.8 or GPT-5.5?

GPT-5.5, with 1.05M tokens against 1M.

How many benchmarks is the Claude Opus 4.8 vs GPT-5.5 comparison based on?

125 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-5.5's on 5. 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-5-5" title="Claude Opus 4.8 vs GPT-5.5, 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-5-5">LLMBenchmark.ai, Claude Opus 4.8 vs GPT-5.5</a></p>