LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

Claude Opus 4.6 vs o4 Mini

Updated October 21, 2026. 80 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace Claude Opus 4.6 Replace o4 Mini
Claude Opus 4.6Anthropic
Coding
66.9
Overall
31.9
Output $/M
$25.00
77
o4 MiniOpenAI
Coding
68.6
Overall
16.7
Output $/M
$4.40
3
vs boards won of 80 shared

Decision reading

Claude Opus 4.6 and o4 Mini are level for coding, 66.9 against 68.6 on the Vibe Coding Index, so price and context decide.

Pick Claude Opus 4.6 when the context window is larger, 1M against 200K tokens, or when what matters is SWE-bench Verified or LiveCodeBench (Vals AI), where it leads. Pick o4 Mini when output tokens cost 5.7x less, or when what matters is Chess puzzles or MedQA (Vals AI), where it leads.

Model AClaude Opus 4.6Anthropic66.9/1002 independent benchmarks$1.56 per solved taskModel Bo4 MiniOpenAI68.6/1002 independent benchmarksNo cost per solved task
Shared results
80
Claude Opus 4.6 only
28
o4 Mini only
14
Like-for-like categories
8 of 9

Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.

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 4.6 66.9 from 2 benchmarks, o4 Mini 68.6 from 2 benchmarks.
Widest category
Web design, 39.6 points apart on average over 6 shared benchmarks.
Widest single gap
ARC-AGI-2: Claude Opus 4.6 69.2% against o4 Mini 6.1%, 63.1 points apart on a 0 to 100 scale.
Output price
Claude Opus 4.6 $25.00 against o4 Mini $4.40 per million output tokens, 5.7x apart.
Context window
Claude Opus 4.6 takes 1M tokens, o4 Mini 200K.

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 4.6Barso4 MiniTies
Coding70
Intelligence20
Agentic work20
Reasoning141
Math60
Knowledge111
Web design60
General291
All shared boards773
CodingLike-for-likeFour of these boards feed the Vibe Coding Index70
Coding benchmarks for Claude Opus 4.6 and o4 Mini
BenchmarkClaude Opus 4.6o4 MiniLeader
Aider polyglotNot measured72.0%Only o4 Mini measured
SWE-bench Verified75.6%64.6%Claude Opus 4.6
SWE-bench MultimodalNot measured33.9%Only o4 Mini measured
SWE-bench Multilingual72.0%Not measuredOnly Claude Opus 4.6 measured
LMArena WebDev1547Not measuredOnly Claude Opus 4.6 measured
LiveCodeBench (Vals AI)84.7%82.2%Claude Opus 4.6
SWE-bench Verified (Vals AI)78.2%Not measuredOnly Claude Opus 4.6 measured
Vibe Code Bench (Vals AI)57.6%Not measuredOnly Claude Opus 4.6 measured
LiveCodeBench (Artificial Analysis run)Not measured85.9%Only o4 Mini measured
Terminal-Bench Hard (Artificial Analysis run)48.5%15.2%Claude Opus 4.6
Terminal-Bench79.8%Not measuredOnly Claude Opus 4.6 measured
FrontierCode26.6%Not measuredOnly Claude Opus 4.6 measured
GSO-Bench41.2%3.6%Claude Opus 4.6
WeirdML78.0%52.6%Claude Opus 4.6
Cybench93.0%Not measuredOnly Claude Opus 4.6 measured
LMArena Text: coding15341368Claude Opus 4.6
LMArena Text: software and IT15271367Claude Opus 4.6
LMArena WebDev: React1539Not measuredOnly Claude Opus 4.6 measured
LMArena WebDev: HTML1559Not measuredOnly Claude Opus 4.6 measured
LMArena WebDev: from an image1543Not measuredOnly Claude Opus 4.6 measured
Neither model has a result on the other 18 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for Claude Opus 4.6 and o4 Mini
BenchmarkClaude Opus 4.6o4 MiniLeader
AA Intelligence Index31.916.7Claude Opus 4.6
Epoch Capabilities Index155.7145.5Claude Opus 4.6
Agentic workLike-for-likePublic boards, shown as published, kept out of the index20
Agentic work benchmarks for Claude Opus 4.6 and o4 Mini
BenchmarkClaude Opus 4.6o4 MiniLeader
Design Arena: agents building web apps1206Not measuredOnly Claude Opus 4.6 measured
Design Arena: agents building full-stack apps1220Not measuredOnly Claude Opus 4.6 measured
Design Arena: agents building mobile apps1207Not measuredOnly Claude Opus 4.6 measured
Design Arena: agents building Android apps1204Not measuredOnly Claude Opus 4.6 measured
Design Arena: agents building games1214Not measuredOnly Claude Opus 4.6 measured
τ²-Bench Telecom92.1%55.6%Claude Opus 4.6
APEX-Agents46.3%Not measuredOnly Claude Opus 4.6 measured
METR time horizons78.9%63.9%Claude Opus 4.6
Remote Labor Index4.2%Not measuredOnly Claude Opus 4.6 measured
DeepResearch Bench55.3%Not measuredOnly Claude Opus 4.6 measured
Vending-Bench 2$8,018Not measuredOnly Claude Opus 4.6 measured
Neither model has a result on the other 28 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index141
Reasoning benchmarks for Claude Opus 4.6 and o4 Mini
BenchmarkClaude Opus 4.6o4 MiniLeader
GPQA Diamond89.7%74.5%Claude Opus 4.6
MMMU Pro (Vals AI)83.9%79.7%Claude Opus 4.6
GPQA Diamond (Artificial Analysis run)89.6%78.4%Claude Opus 4.6
Humanity's Last Exam (Artificial Analysis run)39.9%16.5%Claude Opus 4.6
GPQA Diamond (Epoch AI run)90.5%79.6%Claude Opus 4.6
ARC-AGI94.0%58.7%Claude Opus 4.6
ARC-AGI-269.2%6.1%Claude Opus 4.6
SimpleBench67.6%38.7%Claude Opus 4.6
Chess puzzles21.0%22.0%o4 Mini
Mystery game puzzles25.0%5.0%Claude Opus 4.6
EBR-bench12.7%Not measuredOnly Claude Opus 4.6 measured
VPCTNot measured57.5%Only o4 Mini measured
LMArena Text: hard prompts15251351Claude Opus 4.6
LMArena Text: expert15441343Claude Opus 4.6
LMArena Text: hard prompts in English15301366Claude Opus 4.6
LMArena Vision: diagrams13401197Claude Opus 4.6
LMArena Vision: homework13371234Claude Opus 4.6
Neither model has a result on the other 4 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index60
Math benchmarks for Claude Opus 4.6 and o4 Mini
BenchmarkClaude Opus 4.6o4 MiniLeader
AIME95.6%83.7%Claude Opus 4.6
MATH 500 (Vals AI)Not measured94.2%Only o4 Mini measured
MGSM (Vals AI)Not measured93.4%Only o4 Mini measured
AA Math IndexNot measured90.7Only o4 Mini measured
AIME 2025 (Artificial Analysis run)Not measured90.7%Only o4 Mini measured
OTIS Mock AIME94.4%81.7%Claude Opus 4.6
FrontierMath, tiers 1 to 366.0%36.1%Claude Opus 4.6
FrontierMath, tier 426.8%4.9%Claude Opus 4.6
ProofBench50.0%Not measuredOnly Claude Opus 4.6 measured
MATH level 5Not measured97.8%Only o4 Mini measured
LMArena Text: math15181389Claude Opus 4.6
LMArena Text: mathematical work15251379Claude Opus 4.6
Neither model has a result on the other 1 board here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index111
Knowledge benchmarks for Claude Opus 4.6 and o4 Mini
BenchmarkClaude Opus 4.6o4 MiniLeader
MMLU Pro89.1%80.6%Claude Opus 4.6
LegalBench (Vals AI)85.3%79.2%Claude Opus 4.6
CaseLaw (Vals AI)62.1%Not measuredOnly Claude Opus 4.6 measured
CorpFin (Vals AI)67.0%59.0%Claude Opus 4.6
TaxEval (Vals AI)76.0%74.8%Claude Opus 4.6
MortgageTax (Vals AI)68.5%64.8%Claude Opus 4.6
MedQA (Vals AI)95.4%96.0%o4 Mini
MedCode (Vals AI)49.1%33.8%Claude Opus 4.6
MMLU Pro (Artificial Analysis run)Not measured83.2%Only o4 Mini measured
SimpleQA Verified47.0%19.6%Claude Opus 4.6
GeoBenchNot measured64.0%Only o4 Mini measured
LMArena Text: science15191354Claude Opus 4.6
LMArena Text: medicine and healthcare15041361Claude Opus 4.6
LMArena Text: legal and government15121350Claude Opus 4.6
LMArena Text: business and finance14971325Claude Opus 4.6
Neither model has a result on the other 2 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index60
Web design benchmarks for Claude Opus 4.6 and o4 Mini
BenchmarkClaude Opus 4.6o4 MiniLeader
Design Arena: websites1291985Claude Opus 4.6
Design Arena: UI components1284983Claude Opus 4.6
Design Arena: game dev12781004Claude Opus 4.6
Design Arena: data viz1279990Claude Opus 4.6
Design Arena: 3D1274854Claude Opus 4.6
Design Arena: all code categories1289977Claude Opus 4.6
Design Arena: SVG1227Not measuredOnly Claude Opus 4.6 measured
Design Arena: ASCII art1254Not measuredOnly Claude Opus 4.6 measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index291
General benchmarks for Claude Opus 4.6 and o4 Mini
BenchmarkClaude Opus 4.6o4 MiniLeader
SAGE (Vals AI)51.6%41.1%Claude Opus 4.6
MedScribe (Vals AI)86.7%69.1%Claude Opus 4.6
IFBench53.1%68.7%o4 Mini
Long context reasoning (AA-LCR)78.0%61.0%Claude Opus 4.6
DTBench91.2%77.6%Claude Opus 4.6
LMCA55.8%26.5%Claude Opus 4.6
Fiction.LiveBenchNot measured77.8%Only o4 Mini measured
Creative writing (Lech Mazur)Not measured75.0%Only o4 Mini measured
CL-bench20.7%Not measuredOnly Claude Opus 4.6 measured
Furniture assembly28.3%Not measuredOnly Claude Opus 4.6 measured
CL-bench Life17.0%Not measuredOnly Claude Opus 4.6 measured
LMArena Text15011353Claude Opus 4.6
LMArena Text: instruction following15211320Claude Opus 4.6
LMArena Text: creative writing15051294Claude Opus 4.6
LMArena Vision13161194Claude Opus 4.6
LMArena Text: English15111367Claude Opus 4.6
LMArena Text: non-English14881337Claude Opus 4.6
LMArena Text: Chinese15511354Claude Opus 4.6
LMArena Text: German15001336Claude Opus 4.6
LMArena Text: Japanese14881308Claude Opus 4.6
LMArena Text: Korean14641312Claude Opus 4.6
LMArena Text: French15121364Claude Opus 4.6
LMArena Text: Spanish15101346Claude Opus 4.6
LMArena Text: Russian14971334Claude Opus 4.6
LMArena Text: Polish14991358Claude Opus 4.6
LMArena Text: multi-turn15111350Claude Opus 4.6
LMArena Text: longer queries15181315Claude Opus 4.6
LMArena Text: writing and language15021316Claude Opus 4.6
LMArena Text: entertainment and media14921308Claude Opus 4.6
LMArena Vision: English13241200Claude Opus 4.6
LMArena Vision: Chinese13701218Claude Opus 4.6
LMArena Vision: reading text13261197Claude Opus 4.6
LMArena Vision: humor13141199Claude Opus 4.6
LMArena Vision: captioningNot measured1204Only o4 Mini measured
LMArena Vision: recognizing things12741218Claude Opus 4.6
LMArena Vision: creative writing13311166Claude Opus 4.6
LMArena Document1507Not measuredOnly Claude Opus 4.6 measured
Neither model has a result on the other 9 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Claude Opus 4.6 and o4 Mini
BenchmarkClaude Opus 4.6o4 MiniLeader
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 4.6$0.018Fits in one request
o4 Mini$0.0033Fits in one request

o4 Mini costs less

Repository review

50K input + 3K output tokens

Claude Opus 4.6$0.325Fits in one request
o4 Mini$0.068Fits in one request

o4 Mini costs less

Agent loop

200K input + 10K output tokens

Claude Opus 4.6$1.25Fits in one request
o4 Mini$0.264Exceeds the context window

o4 Mini 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 4.6 and o4 Mini
SpecificationAnthropic Claude Opus 4.6OpenAI o4 Mini
MakerAnthropicOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedFebruary 4, 2026April 16, 2025
Context window1M200K
Input $/Mtok$5.00$1.10
Output $/Mtok$25.00$4.40
Coding index66.968.6
Independent measurements behind it22
Cost per solved task$1.56Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Claude Opus 4.6 better than o4 Mini for coding?

Claude Opus 4.6 and o4 Mini are level for coding, 66.9 against 68.6 on the Vibe Coding Index, so price and context decide. Pick Claude Opus 4.6 when the context window is larger, 1M against 200K tokens, or when what matters is SWE-bench Verified or LiveCodeBench (Vals AI), where it leads. Pick o4 Mini when output tokens cost 5.7x less, or when what matters is Chess puzzles or MedQA (Vals AI), where it leads.

Which is cheaper, Claude Opus 4.6 or o4 Mini?

Claude Opus 4.6 costs $5.00 per million input tokens and $25.00 per million output tokens; o4 Mini costs $1.10 and $4.40. Cost per solved task is known for Claude Opus 4.6 only.

Which has the bigger context window, Claude Opus 4.6 or o4 Mini?

Claude Opus 4.6, with 1M tokens against 200K.

How many benchmarks is the Claude Opus 4.6 vs o4 Mini comparison based on?

80 benchmarks have a result for both models as of October 21, 2026. Claude Opus 4.6's coding index rests on 2 independent benchmarks and o4 Mini's on 2. 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-4-6-vs-o4-mini" title="Claude Opus 4.6 vs o4 Mini, 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-4-6-vs-o4-mini">LLMBenchmark.ai, Claude Opus 4.6 vs o4 Mini</a></p>