Enter opens the highlighted result. Esc closes.

GPT-5 Nano vs o4 Mini

o4 Mini leads GPT-5 Nano on the Artificial Analysis Intelligence Index, 16.7 to 13.0, and GPT-5 Nano costs 11.0x less per output token ($0.40 vs $4.40 per million). Updated October 22, 2026; 84 benchmarks have a result for both.

Replace GPT-5 Nano Replace o4 Mini
GPT-5 NanoOpenAI
Coding
No data
Overall
35.6
Output $/M
$0.40
11
o4 MiniOpenAI
Coding
No data
Overall
43.0
Output $/M
$4.40
72
vs boards won of 84 shared, 1 tied

GPT-5 Nano or o4 Mini: which to pick

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

Pick GPT-5 Nano for output tokens 11.0x cheaper ($0.40 vs $4.40 per million) and a larger context window (400K against 200K tokens).

Model AGPT-5 NanoOpenAINot benchmarkedNo coding benchmark$0.109 per solved taskModel Bo4 MiniOpenAINot benchmarkedNo coding benchmarkNo cost per solved task
Shared results
84
GPT-5 Nano only
3
o4 Mini only
14
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
GPT-5 Nano not benchmarked, o4 Mini not benchmarked.
Widest category
Web design, 10.9 points apart on average over 6 shared benchmarks.
Widest single gap
ARC-AGI: GPT-5 Nano 20.7% against o4 Mini 58.7%, 38.0 points apart on a 0 to 100 scale.
Output price
GPT-5 Nano $0.40 against o4 Mini $4.40 per million output tokens, 11.0x apart.
Context window
GPT-5 Nano takes 400K 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
CategoryGPT-5 NanoBarso4 MiniTies
Coding16
Intelligence02
Agentic work02
Reasoning1141 tie
Math011
Knowledge012
Web design60
General325
All shared boards11721 tie
CodingLike-for-likeSix of these boards feed the Vibe Coding Index16
Coding benchmarks for GPT-5 Nano and o4 Mini
BenchmarkGPT-5 Nanoo4 MiniLeader
Aider polyglotNot measured72.0%Only o4 Mini measured
SWE-bench Verified34.8%64.6%o4 Mini
SWE-bench MultimodalNot measured33.9%Only o4 Mini measured
LiveCodeBench (Vals AI)70.2%82.2%o4 Mini
LiveCodeBench (Artificial Analysis run)78.9%85.9%o4 Mini
Terminal-Bench Hard (Artificial Analysis run)17.4%15.2%GPT-5 Nano
Terminal-Bench21.8%Not measuredOnly GPT-5 Nano measured
GSO-BenchNot measured3.6%Only o4 Mini measured
WeirdML38.1%52.6%o4 Mini
LMArena Text: coding13511368o4 Mini
LMArena Text: software and IT13451367o4 Mini
SciCodeNot measured10.8%Only o4 Mini measured
Neither model has a result on the other 27 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for GPT-5 Nano and o4 Mini
BenchmarkGPT-5 Nanoo4 MiniLeader
AA Intelligence Index13.016.7o4 Mini
Epoch Capabilities Index139.2145.5o4 Mini
Agentic workLike-for-likePublic boards, shown as published, kept out of the index02
Agentic work benchmarks for GPT-5 Nano and o4 Mini
BenchmarkGPT-5 Nanoo4 MiniLeader
τ²-Bench Telecom36.5%55.6%o4 Mini
METR time horizonsNot measured63.9%Only o4 Mini measured
Berkeley Function Calling Leaderboard (BFCL)51.4%53.2%o4 Mini
Neither model has a result on the other 38 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index114
Reasoning benchmarks for GPT-5 Nano and o4 Mini
BenchmarkGPT-5 Nanoo4 MiniLeader
GPQA Diamond63.4%74.5%o4 Mini
MMMU Pro (Vals AI)70.9%79.7%o4 Mini
GPQA Diamond (Artificial Analysis run)67.6%78.4%o4 Mini
Humanity's Last Exam (Artificial Analysis run)9.5%16.5%o4 Mini
GPQA Diamond (Epoch AI run)69.4%79.6%o4 Mini
ARC-AGI20.7%58.7%o4 Mini
ARC-AGI-22.6%6.1%o4 Mini
SimpleBenchNot measured38.7%Only o4 Mini measured
Chess puzzles22.0%22.0%Tie
Mystery game puzzles9.0%5.0%GPT-5 Nano
VPCT37.2%57.5%o4 Mini
LMArena Text: hard prompts13281351o4 Mini
LMArena Text: expert13211343o4 Mini
LMArena Text: hard prompts in English13381366o4 Mini
LMArena Vision: diagrams11411197o4 Mini
LMArena Vision: homework11571234o4 Mini
Kagi LLM Benchmark62.2%67.6%o4 Mini
Neither model has a result on the other 5 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index011
Math benchmarks for GPT-5 Nano and o4 Mini
BenchmarkGPT-5 Nanoo4 MiniLeader
AIME81.2%83.7%o4 Mini
MATH 500 (Vals AI)93.8%94.2%o4 Mini
MGSM (Vals AI)89.3%93.4%o4 Mini
AA Math Index83.790.7o4 Mini
AIME 2025 (Artificial Analysis run)83.7%90.7%o4 Mini
OTIS Mock AIME81.1%81.7%o4 Mini
FrontierMath, tiers 1 to 320.0%36.1%o4 Mini
FrontierMath, tier 42.4%4.9%o4 Mini
ProofBench12.0%Not measuredOnly GPT-5 Nano measured
MATH level 595.2%97.8%o4 Mini
LMArena Text: math13171389o4 Mini
LMArena Text: mathematical work13261379o4 Mini
Neither model has a result on the other 1 board here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index012
Knowledge benchmarks for GPT-5 Nano and o4 Mini
BenchmarkGPT-5 Nanoo4 MiniLeader
MMLU Pro76.1%80.6%o4 Mini
LegalBench (Vals AI)50.1%79.2%o4 Mini
CaseLaw (Vals AI)52.6%Not measuredOnly GPT-5 Nano measured
CorpFin (Vals AI)Not measured59.0%Only o4 Mini measured
TaxEval (Vals AI)67.4%74.8%o4 Mini
MortgageTax (Vals AI)53.6%64.8%o4 Mini
MedQA (Vals AI)93.3%96.0%o4 Mini
MedCode (Vals AI)30.4%33.8%o4 Mini
MMLU Pro (Artificial Analysis run)78.0%83.2%o4 Mini
SimpleQA Verified11.7%19.6%o4 Mini
GeoBenchNot measured64.0%Only o4 Mini measured
LMArena Text: science13181354o4 Mini
LMArena Text: medicine and healthcare13351361o4 Mini
LMArena Text: legal and government12981350o4 Mini
LMArena Text: business and finance13111325o4 Mini
FACTS Benchmark SuiteNot measured28.1%Only o4 Mini measured
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 GPT-5 Nano and o4 Mini
BenchmarkGPT-5 Nanoo4 MiniLeader
Design Arena: websites1101985GPT-5 Nano
Design Arena: UI components1064983GPT-5 Nano
Design Arena: game dev10441003GPT-5 Nano
Design Arena: data viz1056991GPT-5 Nano
Design Arena: 3D965854GPT-5 Nano
Design Arena: all code categories1084976GPT-5 Nano
Neither model has a result on the other 3 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index325
General benchmarks for GPT-5 Nano and o4 Mini
BenchmarkGPT-5 Nanoo4 MiniLeader
SAGE (Vals AI)30.4%41.1%o4 Mini
MedScribe (Vals AI)72.9%69.1%GPT-5 Nano
IFBench67.6%68.7%o4 Mini
Long context reasoning (AA-LCR)45.0%61.0%o4 Mini
DTBench62.7%77.6%o4 Mini
LMCA7.9%26.5%o4 Mini
Fiction.LiveBench44.4%77.8%o4 Mini
Creative writing (Lech Mazur)Not measured75.0%Only o4 Mini measured
LMArena Text13201353o4 Mini
LMArena Text: instruction following13061320o4 Mini
LMArena Text: creative writing12491294o4 Mini
LMArena Vision11591194o4 Mini
LMArena Text: English13211367o4 Mini
LMArena Text: non-English13131337o4 Mini
LMArena Text: Chinese13561354GPT-5 Nano
LMArena Text: German13271336o4 Mini
LMArena Text: Japanese12271308o4 Mini
LMArena Text: Korean12691312o4 Mini
LMArena Text: FrenchNot measured1364Only o4 Mini measured
LMArena Text: Spanish13601346GPT-5 Nano
LMArena Text: Russian12961334o4 Mini
LMArena Text: Polish13141358o4 Mini
LMArena Text: multi-turn13111350o4 Mini
LMArena Text: longer queries13121315o4 Mini
LMArena Text: writing and language12831316o4 Mini
LMArena Text: entertainment and media12681308o4 Mini
LMArena Vision: English11701200o4 Mini
LMArena Vision: Chinese11761218o4 Mini
LMArena Vision: reading text11521197o4 Mini
LMArena Vision: humor11101199o4 Mini
LMArena Vision: captioningNot measured1204Only o4 Mini measured
LMArena Vision: recognizing thingsNot measured1218Only o4 Mini measured
LMArena Vision: creative writingNot measured1166Only o4 Mini measured
Neither model has a result on the other 13 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GPT-5 Nano and o4 Mini
BenchmarkGPT-5 Nanoo4 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

GPT-5 Nano$0.0003Fits in one request
o4 Mini$0.0033Fits in one request

GPT-5 Nano costs less

Repository review

50K input + 3K output tokens

GPT-5 Nano$0.0037Fits in one request
o4 Mini$0.068Fits in one request

GPT-5 Nano costs less

Agent loop

200K input + 10K output tokens

GPT-5 Nano$0.014Fits in one request
o4 Mini$0.264Exceeds the context window

GPT-5 Nano 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-5 Nano and o4 Mini
SpecificationOpenAI GPT-5 NanoOpenAI o4 Mini
MakerOpenAIOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedAugust 7, 2025April 16, 2025
Context window400K200K
Input $/Mtok$0.05$1.10
Output $/Mtok$0.40$4.40
Coding indexNot benchmarkedNot benchmarked
Independent measurements behind it00
Cost per solved task$0.109Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-5 Nano better than o4 Mini for coding?

Neither model has a coding score yet, so the two cannot be ranked on coding. Pick GPT-5 Nano for output tokens 11.0x cheaper ($0.40 vs $4.40 per million) and a larger context window (400K against 200K tokens).

Which is cheaper, GPT-5 Nano or o4 Mini?

GPT-5 Nano costs $0.05 per million input tokens and $0.40 per million output tokens; o4 Mini costs $1.10 and $4.40. Cost per solved task is known for GPT-5 Nano only.

Which has the bigger context window, GPT-5 Nano or o4 Mini?

GPT-5 Nano, with 400K tokens against 200K.

How many benchmarks is the GPT-5 Nano vs o4 Mini comparison based on?

84 benchmarks have a result for both models as of October 22, 2026. GPT-5 Nano's coding index rests on 0 independent benchmarks and o4 Mini's on 0. Where only one model has a result, the page says so instead of naming a winner.

Compare next