Enter opens the highlighted result. Esc closes.

GPT-5 Mini vs o4 Mini

GPT-5 Mini leads o4 Mini on the Artificial Analysis Intelligence Index, 20.6 to 16.7, and GPT-5 Mini costs 2.2x less per output token ($2.00 vs $4.40 per million). Updated October 22, 2026; 91 benchmarks have a result for both.

Replace GPT-5 Mini Replace o4 Mini
GPT-5 MiniOpenAI
Coding
34.2
Overall
43.7
Output $/M
$2.00
75
o4 MiniOpenAI
Coding
No data
Overall
43.0
Output $/M
$4.40
14
vs boards won of 91 shared, 2 tied

GPT-5 Mini or o4 Mini: which to pick

Only GPT-5 Mini has a coding score, 34.2 on the Vibe Coding Index; o4 Mini has none, so the two cannot be ranked on coding.

Pick o4 Mini for work measured by SWE-bench Verified or LiveCodeBench (Artificial Analysis run), where it leads.

Model AGPT-5 MiniOpenAI34.2/1002 independent benchmarks$0.08 per solved taskModel Bo4 MiniOpenAINot benchmarkedNo coding benchmarkNo cost per solved task
Shared results
91
GPT-5 Mini only
10
o4 Mini only
7
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 Mini 34.2 from 2 benchmarks, o4 Mini not benchmarked.
Widest category
Web design, 17.6 points apart on average over 6 shared benchmarks.
Widest single gap
Design Arena: 3D: GPT-5 Mini 1036 against o4 Mini 854, 22.8 points apart on a 0 to 100 scale.
Output price
GPT-5 Mini $2.00 against o4 Mini $4.40 per million output tokens, 2.2x apart.
Context window
GPT-5 Mini 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 MiniBarso4 MiniTies
Coding52
Intelligence11
Agentic work20
Reasoning106
Math722 ties
Knowledge140
Web design60
General303
All shared boards75142 ties
CodingLike-for-likeSix of these boards feed the Vibe Coding Index52
Coding benchmarks for GPT-5 Mini and o4 Mini
BenchmarkGPT-5 Minio4 MiniLeader
Aider polyglotNot measured72.0%Only o4 Mini measured
SWE-bench Verified59.8%64.6%o4 Mini
SWE-bench MultimodalNot measured33.9%Only o4 Mini measured
SWE-bench Multilingual39.7%Not measuredOnly GPT-5 Mini measured
LiveCodeBench (Vals AI)86.6%82.2%GPT-5 Mini
SWE-bench Verified (Vals AI)60.8%Not measuredOnly GPT-5 Mini measured
Vibe Code Bench (Vals AI)14.2%Not measuredOnly GPT-5 Mini measured
LiveCodeBench (Artificial Analysis run)83.8%85.9%o4 Mini
SciCode39.0%Not measuredOnly GPT-5 Mini measured
Terminal-Bench Hard (Artificial Analysis run)33.3%15.2%GPT-5 Mini
Terminal-Bench34.8%Not measuredOnly GPT-5 Mini measured
GSO-BenchNot measured3.6%Only o4 Mini measured
WeirdML52.7%52.6%GPT-5 Mini
LMArena Text: coding14061368GPT-5 Mini
LMArena Text: software and IT13941367GPT-5 Mini
SciCodeNot measured10.8%Only o4 Mini measured
Neither model has a result on the other 23 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index11
Intelligence benchmarks for GPT-5 Mini and o4 Mini
BenchmarkGPT-5 Minio4 MiniLeader
AA Intelligence Index20.616.7GPT-5 Mini
Epoch Capabilities Index145.3145.5o4 Mini
Agentic workLike-for-likePublic boards, shown as published, kept out of the index20
Agentic work benchmarks for GPT-5 Mini and o4 Mini
BenchmarkGPT-5 Minio4 MiniLeader
τ²-Bench Telecom71.1%55.6%GPT-5 Mini
METR time horizonsNot measured63.9%Only o4 Mini measured
Vending-Bench 2-$31Not measuredOnly GPT-5 Mini measured
Berkeley Function Calling Leaderboard (BFCL)55.5%53.2%GPT-5 Mini
Neither model has a result on the other 37 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index106
Reasoning benchmarks for GPT-5 Mini and o4 Mini
BenchmarkGPT-5 Minio4 MiniLeader
GPQA Diamond80.3%74.5%GPT-5 Mini
MMMU Pro (Vals AI)78.9%79.7%o4 Mini
GPQA Diamond (Artificial Analysis run)82.8%78.4%GPT-5 Mini
Humanity's Last Exam (Artificial Analysis run)21.5%16.5%GPT-5 Mini
GPQA Diamond (Epoch AI run)75.0%79.6%o4 Mini
ARC-AGI54.3%58.7%o4 Mini
ARC-AGI-24.4%6.1%o4 Mini
SimpleBenchNot measured38.7%Only o4 Mini measured
Chess puzzles24.0%22.0%GPT-5 Mini
Mystery game puzzles10.0%5.0%GPT-5 Mini
VPCT40.2%57.5%o4 Mini
LMArena Text: hard prompts13801351GPT-5 Mini
LMArena Text: expert13791343GPT-5 Mini
LMArena Text: hard prompts in English13831366GPT-5 Mini
LMArena Vision: diagrams12021197GPT-5 Mini
LMArena Vision: homework12101234o4 Mini
Kagi LLM Benchmark70.3%67.6%GPT-5 Mini
Neither model has a result on the other 5 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index72
Math benchmarks for GPT-5 Mini and o4 Mini
BenchmarkGPT-5 Minio4 MiniLeader
AIME91.5%83.7%GPT-5 Mini
MATH 500 (Vals AI)94.8%94.2%GPT-5 Mini
MGSM (Vals AI)92.6%93.4%o4 Mini
AA Math Index90.790.7Tie
AIME 2025 (Artificial Analysis run)90.7%90.7%Tie
OTIS Mock AIME86.7%81.7%GPT-5 Mini
FrontierMath, tiers 1 to 346.7%36.1%GPT-5 Mini
FrontierMath, tier 412.2%4.9%GPT-5 Mini
ProofBench9.0%Not measuredOnly GPT-5 Mini measured
MATH level 597.8%97.8%GPT-5 Mini
LMArena Text: math13781389o4 Mini
LMArena Text: mathematical work13801379GPT-5 Mini
Neither model has a result on the other 1 board here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index140
Knowledge benchmarks for GPT-5 Mini and o4 Mini
BenchmarkGPT-5 Minio4 MiniLeader
MMLU Pro82.2%80.6%GPT-5 Mini
LegalBench (Vals AI)81.8%79.2%GPT-5 Mini
CaseLaw (Vals AI)68.5%Not measuredOnly GPT-5 Mini measured
CorpFin (Vals AI)60.2%59.0%GPT-5 Mini
TaxEval (Vals AI)75.2%74.8%GPT-5 Mini
MortgageTax (Vals AI)66.9%64.8%GPT-5 Mini
MedQA (Vals AI)96.1%96.0%GPT-5 Mini
MedCode (Vals AI)43.0%33.8%GPT-5 Mini
MMLU Pro (Artificial Analysis run)83.7%83.2%GPT-5 Mini
SimpleQA Verified21.6%19.6%GPT-5 Mini
GeoBenchNot measured64.0%Only o4 Mini measured
LMArena Text: science13711354GPT-5 Mini
LMArena Text: medicine and healthcare13801361GPT-5 Mini
LMArena Text: legal and government13751350GPT-5 Mini
LMArena Text: business and finance13581325GPT-5 Mini
FACTS Benchmark Suite38.6%28.1%GPT-5 Mini
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 Mini and o4 Mini
BenchmarkGPT-5 Minio4 MiniLeader
Design Arena: websites1125985GPT-5 Mini
Design Arena: UI components1103983GPT-5 Mini
Design Arena: game dev11291003GPT-5 Mini
Design Arena: data viz1131991GPT-5 Mini
Design Arena: 3D1036854GPT-5 Mini
Design Arena: all code categories1114976GPT-5 Mini
Design Arena: SVG1091Not measuredOnly GPT-5 Mini measured
Design Arena: ASCII art1125Not measuredOnly GPT-5 Mini measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index303
General benchmarks for GPT-5 Mini and o4 Mini
BenchmarkGPT-5 Minio4 MiniLeader
SAGE (Vals AI)43.0%41.1%GPT-5 Mini
MedScribe (Vals AI)80.6%69.1%GPT-5 Mini
IFBench75.4%68.7%GPT-5 Mini
Long context reasoning (AA-LCR)72.3%61.0%GPT-5 Mini
DTBench80.5%77.6%GPT-5 Mini
LMCA34.2%26.5%GPT-5 Mini
Fiction.LiveBench69.4%77.8%o4 Mini
Creative writing (Lech Mazur)83.1%75.0%GPT-5 Mini
LMArena Text13731353GPT-5 Mini
LMArena Text: instruction following13571320GPT-5 Mini
LMArena Text: creative writing13251294GPT-5 Mini
LMArena Vision12021194GPT-5 Mini
LMArena Text: English13791367GPT-5 Mini
LMArena Text: non-English13631337GPT-5 Mini
LMArena Text: Chinese13861354GPT-5 Mini
LMArena Text: German13661336GPT-5 Mini
LMArena Text: Japanese13411308GPT-5 Mini
LMArena Text: Korean13081312o4 Mini
LMArena Text: French13861364GPT-5 Mini
LMArena Text: Spanish13541346GPT-5 Mini
LMArena Text: Russian13621334GPT-5 Mini
LMArena Text: Polish13781358GPT-5 Mini
LMArena Text: multi-turn13631350GPT-5 Mini
LMArena Text: longer queries13551315GPT-5 Mini
LMArena Text: writing and language13441316GPT-5 Mini
LMArena Text: entertainment and media13401308GPT-5 Mini
LMArena Vision: English12161200GPT-5 Mini
LMArena Vision: Chinese12141218o4 Mini
LMArena Vision: reading text11991197GPT-5 Mini
LMArena Vision: humor12051199GPT-5 Mini
LMArena Vision: captioning12171204GPT-5 Mini
LMArena Vision: recognizing things12261218GPT-5 Mini
LMArena Vision: creative writing11811166GPT-5 Mini
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 Mini and o4 Mini
BenchmarkGPT-5 Minio4 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 Mini$0.0013Fits in one request
o4 Mini$0.0033Fits in one request

GPT-5 Mini costs less

Repository review

50K input + 3K output tokens

GPT-5 Mini$0.018Fits in one request
o4 Mini$0.068Fits in one request

GPT-5 Mini costs less

Agent loop

200K input + 10K output tokens

GPT-5 Mini$0.070Fits in one request
o4 Mini$0.264Exceeds the context window

GPT-5 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-5 Mini and o4 Mini
SpecificationOpenAI GPT-5 MiniOpenAI o4 Mini
MakerOpenAIOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedAugust 7, 2025April 16, 2025
Context window400K200K
Input $/Mtok$0.25$1.10
Output $/Mtok$2.00$4.40
Coding index34.2Not benchmarked
Independent measurements behind it20
Cost per solved task$0.08Not computable
SourceOpenRouter listingOpenRouter listing

Questions

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

Only GPT-5 Mini has a coding score, 34.2 on the Vibe Coding Index; o4 Mini has none, so the two cannot be ranked on coding. Pick o4 Mini for work measured by SWE-bench Verified or LiveCodeBench (Artificial Analysis run), where it leads.

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

GPT-5 Mini costs $0.25 per million input tokens and $2.00 per million output tokens; o4 Mini costs $1.10 and $4.40. Cost per solved task is known for GPT-5 Mini only.

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

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

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

91 benchmarks have a result for both models as of October 22, 2026. GPT-5 Mini's coding index rests on 2 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