Enter opens the highlighted result. Esc closes.

GPT-4o-mini vs o3

o3 leads GPT-4o-mini on the Artificial Analysis Intelligence Index, 20.2 to 6.7, and GPT-4o-mini costs 13.3x less per output token ($0.60 vs $8.00 per million). Updated October 21, 2026; 25 benchmarks have a result for both.

Replace GPT-4o-mini Replace o3
GPT-4o-miniOpenAI
Coding
No data
Overall
6.7
Output $/M
$0.60
0
o3OpenAI
Coding
No data
Overall
20.2
Output $/M
$8.00
25
vs boards won of 25 shared

GPT-4o-mini or o3: which to pick

Neither model has a coding score yet, so the two cannot be ranked on coding.

Pick GPT-4o-mini for output tokens 13.3x cheaper ($0.60 vs $8.00 per million) and work measured by LiveCodeBench (Vals AI) or LiveCodeBench (Artificial Analysis run), where it leads.

Model AGPT-4o-miniOpenAINot benchmarkedNo coding benchmarkNo cost per solved taskModel Bo3OpenAINot benchmarkedNo coding benchmark$0.571 per solved task
Shared results
25
GPT-4o-mini only
4
o3 only
77
Like-for-like categories
6 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
GPT-4o-mini not benchmarked, o3 not benchmarked.
Widest category
Math, 49.7 points apart on average over 5 shared benchmarks.
Widest single gap
AIME: GPT-4o-mini 11.5% against o3 85.3%, 73.8 points apart on a 0 to 100 scale.
Output price
GPT-4o-mini $0.60 against o3 $8.00 per million output tokens, 13.3x apart.
Context window
GPT-4o-mini takes 128K tokens, o3 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
CategoryGPT-4o-miniBarso3Ties
Coding03
Intelligence02
Reasoning06
Math05
Knowledge06
General03
All shared boards025
CodingLike-for-likeSix of these boards feed the Vibe Coding Index03
Coding benchmarks for GPT-4o-mini and o3
BenchmarkGPT-4o-minio3Leader
Aider polyglotNot measured81.3%Only o3 measured
SWE-bench VerifiedNot measured58.4%Only o3 measured
SWE-bench MultimodalNot measured36.0%Only o3 measured
AA Coding Index11.4Not measuredOnly GPT-4o-mini measured
LiveCodeBench (Vals AI)26.4%83.9%o3
LiveCodeBench (Artificial Analysis run)23.4%80.8%o3
Terminal-Bench Hard (Artificial Analysis run)Not measured37.1%Only o3 measured
GSO-BenchNot measured8.8%Only o3 measured
WeirdMLNot measured52.4%Only o3 measured
ALE-BenchNot measured933.5Only o3 measured
LMArena Text: codingNot measured1407Only o3 measured
LMArena Text: software and ITNot measured1415Only o3 measured
SciCode1.5%9.2%o3
Neither model has a result on the other 26 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for GPT-4o-mini and o3
BenchmarkGPT-4o-minio3Leader
AA Intelligence Index6.720.2o3
Epoch Capabilities Index126.5146.7o3
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
Agentic work benchmarks for GPT-4o-mini and o3
BenchmarkGPT-4o-minio3Leader
τ²-Bench TelecomNot measured80.7%Only o3 measured
METR time horizonsNot measured65.4%Only o3 measured
GDPvalNot measured30.8%Only o3 measured
DeepResearch BenchNot measured45.2%Only o3 measured
OSWorldNot measured23.0%Only o3 measured
Berkeley Function Calling Leaderboard (BFCL)Not measured63.0%Only o3 measured
BrowseComp-PlusNot measured65.9%Only o3 measured
Neither model has a result on the other 34 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index06
Reasoning benchmarks for GPT-4o-mini and o3
BenchmarkGPT-4o-minio3Leader
GPQA Diamond44.2%84.1%o3
MMMU Pro (Vals AI)56.6%80.4%o3
GPQA Diamond (Artificial Analysis run)42.6%82.7%o3
Humanity's Last Exam (Artificial Analysis run)4.2%20.1%o3
GPQA Diamond (Epoch AI run)Not measured81.8%Only o3 measured
ARC-AGINot measured60.8%Only o3 measured
ARC-AGI-20.0%6.5%o3
SimpleBenchNot measured53.1%Only o3 measured
Chess puzzlesNot measured37.0%Only o3 measured
Mystery game puzzlesNot measured29.0%Only o3 measured
VPCTNot measured52.0%Only o3 measured
LMArena Text: hard promptsNot measured1402Only o3 measured
LMArena Text: expertNot measured1402Only o3 measured
LMArena Text: hard prompts in EnglishNot measured1411Only o3 measured
LMArena Vision: diagramsNot measured1220Only o3 measured
LMArena Vision: homeworkNot measured1234Only o3 measured
Kagi LLM Benchmark28.8%67.6%o3
Neither model has a result on the other 5 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index05
Math benchmarks for GPT-4o-mini and o3
BenchmarkGPT-4o-minio3Leader
AIME11.5%85.3%o3
MATH 500 (Vals AI)72.6%94.6%o3
MGSM (Vals AI)86.2%91.8%o3
AA Math Index14.788.3o3
AIME 2025 (Artificial Analysis run)14.7%88.3%o3
OTIS Mock AIMENot measured84.4%Only o3 measured
FrontierMath, tiers 1 to 3Not measured33.3%Only o3 measured
MATH level 5Not measured97.8%Only o3 measured
GSM8K91.3%Not measuredOnly GPT-4o-mini measured
LMArena Text: mathNot measured1426Only o3 measured
LMArena Text: mathematical workNot measured1423Only o3 measured
Neither model has a result on the other 2 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index06
Knowledge benchmarks for GPT-4o-mini and o3
BenchmarkGPT-4o-minio3Leader
MMLU Pro62.7%85.6%o3
LegalBench (Vals AI)Not measured83.8%Only o3 measured
CorpFin (Vals AI)45.5%59.7%o3
TaxEval (Vals AI)60.5%74.6%o3
MortgageTax (Vals AI)54.5%65.7%o3
MedQA (Vals AI)72.4%96.1%o3
MedCode (Vals AI)Not measured47.3%Only o3 measured
MMLU Pro (Artificial Analysis run)64.8%85.3%o3
SimpleQA VerifiedNot measured49.4%Only o3 measured
MMLU81.8%Not measuredOnly GPT-4o-mini measured
GeoBenchNot measured74.0%Only o3 measured
LMArena Text: scienceNot measured1429Only o3 measured
LMArena Text: medicine and healthcareNot measured1448Only o3 measured
LMArena Text: legal and governmentNot measured1428Only o3 measured
LMArena Text: business and financeNot measured1393Only o3 measured
FACTS Benchmark SuiteNot measured44.4%Only o3 measured
Neither model has a result on the other 2 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index5 boards, none shared
Web design benchmarks for GPT-4o-mini and o3
BenchmarkGPT-4o-minio3Leader
Design Arena: websitesNot measured1036Only o3 measured
Design Arena: UI componentsNot measured1017Only o3 measured
Design Arena: game devNot measured1034Only o3 measured
Design Arena: data vizNot measured1200Only o3 measured
Design Arena: all code categoriesNot measured1020Only o3 measured
Neither model has a result on the other 4 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index03
General benchmarks for GPT-4o-mini and o3
BenchmarkGPT-4o-minio3Leader
SAGE (Vals AI)Not measured41.8%Only o3 measured
MedScribe (Vals AI)Not measured76.7%Only o3 measured
IFBench31.0%71.4%o3
Long context reasoning (AA-LCR)Not measured74.7%Only o3 measured
DTBench54.4%84.8%o3
LMCA10.4%39.7%o3
Fiction.LiveBenchNot measured88.9%Only o3 measured
Creative writing (Lech Mazur)Not measured83.9%Only o3 measured
CL-benchNot measured17.8%Only o3 measured
PIQA88.7%Not measuredOnly GPT-4o-mini measured
LMArena TextNot measured1410Only o3 measured
LMArena Text: instruction followingNot measured1368Only o3 measured
LMArena Text: creative writingNot measured1359Only o3 measured
LMArena VisionNot measured1214Only o3 measured
LMArena Text: EnglishNot measured1417Only o3 measured
LMArena Text: non-EnglishNot measured1401Only o3 measured
LMArena Text: ChineseNot measured1437Only o3 measured
LMArena Text: GermanNot measured1420Only o3 measured
LMArena Text: JapaneseNot measured1403Only o3 measured
LMArena Text: KoreanNot measured1370Only o3 measured
LMArena Text: FrenchNot measured1430Only o3 measured
LMArena Text: SpanishNot measured1395Only o3 measured
LMArena Text: RussianNot measured1406Only o3 measured
LMArena Text: PolishNot measured1438Only o3 measured
LMArena Text: multi-turnNot measured1405Only o3 measured
LMArena Text: longer queriesNot measured1372Only o3 measured
LMArena Text: writing and languageNot measured1372Only o3 measured
LMArena Text: entertainment and mediaNot measured1373Only o3 measured
LMArena Vision: EnglishNot measured1221Only o3 measured
LMArena Vision: ChineseNot measured1237Only o3 measured
LMArena Vision: reading textNot measured1218Only o3 measured
LMArena Vision: humorNot measured1222Only o3 measured
LMArena Vision: captioningNot measured1210Only o3 measured
LMArena Vision: recognizing thingsNot measured1234Only o3 measured
LMArena Vision: creative writingNot measured1201Only o3 measured
Neither model has a result on the other 11 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GPT-4o-mini and o3
BenchmarkGPT-4o-minio3Leader
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

GPT-4o-mini$0.0004Fits in one request
o3$0.0060Fits in one request

GPT-4o-mini costs less

Repository review

50K input + 3K output tokens

GPT-4o-mini$0.0093Fits in one request
o3$0.124Fits in one request

GPT-4o-mini costs less

Agent loop

200K input + 10K output tokens

GPT-4o-mini$0.036Exceeds the context window
o3$0.480Exceeds the context window

GPT-4o-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 GPT-4o-mini and o3
SpecificationOpenAI GPT-4o-miniOpenAI o3
MakerOpenAIOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedJuly 18, 2024April 16, 2025
Context window128K200K
Input $/Mtok$0.15$2.00
Output $/Mtok$0.60$8.00
Coding indexNot benchmarkedNot benchmarked
Independent measurements behind it00
Cost per solved taskNot computable$0.571
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-4o-mini better than o3 for coding?

Neither model has a coding score yet, so the two cannot be ranked on coding. Pick GPT-4o-mini for output tokens 13.3x cheaper ($0.60 vs $8.00 per million) and work measured by LiveCodeBench (Vals AI) or LiveCodeBench (Artificial Analysis run), where it leads.

Which is cheaper, GPT-4o-mini or o3?

GPT-4o-mini costs $0.15 per million input tokens and $0.60 per million output tokens; o3 costs $2.00 and $8.00. Cost per solved task is known for o3 only.

Which has the bigger context window, GPT-4o-mini or o3?

o3, with 200K tokens against 128K.

How many benchmarks is the GPT-4o-mini vs o3 comparison based on?

25 benchmarks have a result for both models as of October 21, 2026. GPT-4o-mini's coding index rests on 0 independent benchmarks and o3's on 0. Where only one model has a result, the page says so instead of naming a winner.

Compare next