LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

GPT-4.1 Mini vs GPT-4.1 Nano

Updated October 21, 2026. 69 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace GPT-4.1 Mini Replace GPT-4.1 Nano
GPT-4.1 MiniOpenAI
Coding
28.5
Overall
10.2
Output $/M
$1.60
67
GPT-4.1 NanoOpenAI
Coding
8.9
Overall
7.8
Output $/M
$0.40
1
vs boards won of 69 shared, 1 tied

Decision reading

Pick GPT-4.1 Mini for coding: it scores 28.5 on the Vibe Coding Index against 8.9 for GPT-4.1 Nano.

Pick GPT-4.1 Nano when output tokens cost 4.0x less, or when the benchmark that matters is Design Arena: 3D, where it leads.

Model AGPT-4.1 MiniOpenAI28.5/1002 independent benchmarks$1.818 per solved taskModel BGPT-4.1 NanoOpenAI8.9/1001 independent benchmarkNo cost per solved task
Shared results
69
GPT-4.1 Mini only
15
GPT-4.1 Nano only
0
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
GPT-4.1 Mini 28.5 from 2 benchmarks, GPT-4.1 Nano 8.9 from 1 benchmark.
Widest category
Agentic work, 35.7 points apart on average over 1 shared benchmark.
Widest single gap
τ²-Bench Telecom: GPT-4.1 Mini 52.9% against GPT-4.1 Nano 17.3%, 35.7 points apart on a 0 to 100 scale.
Output price
GPT-4.1 Mini $1.60 against GPT-4.1 Nano $0.40 per million output tokens, 4.0x apart.
Context window
Both take 1.05M tokens.

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-4.1 MiniBarsGPT-4.1 NanoTies
Coding80
Intelligence20
Agentic work10
Reasoning1001 tie
Math90
Knowledge120
Web design51
General200
All shared boards6711 tie
CodingLike-for-likeFour of these boards feed the Vibe Coding Index80
Coding benchmarks for GPT-4.1 Mini and GPT-4.1 Nano
BenchmarkGPT-4.1 MiniGPT-4.1 NanoLeader
Aider polyglot32.4%8.9%GPT-4.1 Mini
SWE-bench Verified23.9%Not measuredOnly GPT-4.1 Mini measured
AA Coding Index20.211.1GPT-4.1 Mini
LiveCodeBench (Vals AI)58.2%42.7%GPT-4.1 Mini
LiveCodeBench (Artificial Analysis run)48.3%32.6%GPT-4.1 Mini
Terminal-Bench Hard (Artificial Analysis run)7.6%3.8%GPT-4.1 Mini
WeirdML37.6%19.0%GPT-4.1 Mini
LMArena Text: coding13671306GPT-4.1 Mini
LMArena Text: software and IT13601305GPT-4.1 Mini
Neither model has a result on the other 29 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for GPT-4.1 Mini and GPT-4.1 Nano
BenchmarkGPT-4.1 MiniGPT-4.1 NanoLeader
AA Intelligence Index10.27.8GPT-4.1 Mini
Epoch Capabilities Index134.9129.5GPT-4.1 Mini
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
Agentic work benchmarks for GPT-4.1 Mini and GPT-4.1 Nano
BenchmarkGPT-4.1 MiniGPT-4.1 NanoLeader
τ²-Bench Telecom52.9%17.3%GPT-4.1 Mini
Neither model has a result on the other 38 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index100
Reasoning benchmarks for GPT-4.1 Mini and GPT-4.1 Nano
BenchmarkGPT-4.1 MiniGPT-4.1 NanoLeader
GPQA Diamond67.9%50.8%GPT-4.1 Mini
MMMU Pro (Vals AI)70.5%55.0%GPT-4.1 Mini
GPQA Diamond (Artificial Analysis run)66.4%51.2%GPT-4.1 Mini
Humanity's Last Exam (Artificial Analysis run)5.0%3.8%GPT-4.1 Mini
GPQA Diamond (Epoch AI run)65.8%48.9%GPT-4.1 Mini
ARC-AGI3.5%0.0%GPT-4.1 Mini
ARC-AGI-20.0%0.0%Tie
Chess puzzles1.0%0.0%GPT-4.1 Mini
Mystery game puzzles7.0%Not measuredOnly GPT-4.1 Mini measured
LMArena Text: hard prompts13491286GPT-4.1 Mini
LMArena Text: expert13381271GPT-4.1 Mini
LMArena Text: hard prompts in English13621293GPT-4.1 Mini
LMArena Vision: diagrams1182Not measuredOnly GPT-4.1 Mini measured
LMArena Vision: homework1214Not measuredOnly GPT-4.1 Mini measured
Neither model has a result on the other 7 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index90
Math benchmarks for GPT-4.1 Mini and GPT-4.1 Nano
BenchmarkGPT-4.1 MiniGPT-4.1 NanoLeader
AIME49.4%26.5%GPT-4.1 Mini
MATH 500 (Vals AI)88.0%80.2%GPT-4.1 Mini
MGSM (Vals AI)87.8%69.3%GPT-4.1 Mini
AA Math Index46.324.0GPT-4.1 Mini
AIME 2025 (Artificial Analysis run)46.3%24.0%GPT-4.1 Mini
OTIS Mock AIME44.7%28.9%GPT-4.1 Mini
FrontierMath, tiers 1 to 36.7%Not measuredOnly GPT-4.1 Mini measured
MATH level 587.3%70.0%GPT-4.1 Mini
LMArena Text: math13431274GPT-4.1 Mini
LMArena Text: mathematical work13471274GPT-4.1 Mini
Neither model has a result on the other 3 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index120
Knowledge benchmarks for GPT-4.1 Mini and GPT-4.1 Nano
BenchmarkGPT-4.1 MiniGPT-4.1 NanoLeader
MMLU Pro77.2%63.5%GPT-4.1 Mini
LegalBench (Vals AI)78.0%61.1%GPT-4.1 Mini
CorpFin (Vals AI)57.9%42.1%GPT-4.1 Mini
TaxEval (Vals AI)71.9%60.8%GPT-4.1 Mini
MortgageTax (Vals AI)65.5%52.8%GPT-4.1 Mini
MedQA (Vals AI)84.6%68.2%GPT-4.1 Mini
MMLU Pro (Artificial Analysis run)78.1%65.7%GPT-4.1 Mini
SimpleQA Verified12.7%6.0%GPT-4.1 Mini
LMArena Text: science13281297GPT-4.1 Mini
LMArena Text: medicine and healthcare13241276GPT-4.1 Mini
LMArena Text: legal and government13431300GPT-4.1 Mini
LMArena Text: business and finance13221267GPT-4.1 Mini
Neither model has a result on the other 5 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index51
Web design benchmarks for GPT-4.1 Mini and GPT-4.1 Nano
BenchmarkGPT-4.1 MiniGPT-4.1 NanoLeader
Design Arena: websites997973GPT-4.1 Mini
Design Arena: UI components961916GPT-4.1 Mini
Design Arena: game dev1071971GPT-4.1 Mini
Design Arena: data viz1034891GPT-4.1 Mini
Design Arena: 3D837925GPT-4.1 Nano
Design Arena: all code categories994963GPT-4.1 Mini
Neither model has a result on the other 3 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index200
General benchmarks for GPT-4.1 Mini and GPT-4.1 Nano
BenchmarkGPT-4.1 MiniGPT-4.1 NanoLeader
IFBench38.3%32.0%GPT-4.1 Mini
Long context reasoning (AA-LCR)44.0%20.3%GPT-4.1 Mini
DTBench68.8%52.5%GPT-4.1 Mini
LMCA21.1%5.5%GPT-4.1 Mini
Fiction.LiveBench44.4%25.0%GPT-4.1 Mini
LMArena Text13411285GPT-4.1 Mini
LMArena Text: instruction following13331266GPT-4.1 Mini
LMArena Text: creative writing12991260GPT-4.1 Mini
LMArena Vision11811063GPT-4.1 Mini
LMArena Text: English13561303GPT-4.1 Mini
LMArena Text: non-English13181260GPT-4.1 Mini
LMArena Text: Chinese13291270GPT-4.1 Mini
LMArena Text: German13501288GPT-4.1 Mini
LMArena Text: Japanese12901198GPT-4.1 Mini
LMArena Text: Korean1298Not measuredOnly GPT-4.1 Mini measured
LMArena Text: French1358Not measuredOnly GPT-4.1 Mini measured
LMArena Text: Spanish1319Not measuredOnly GPT-4.1 Mini measured
LMArena Text: Russian13241261GPT-4.1 Mini
LMArena Text: Polish1326Not measuredOnly GPT-4.1 Mini measured
LMArena Text: multi-turn13541277GPT-4.1 Mini
LMArena Text: longer queries13441283GPT-4.1 Mini
LMArena Text: writing and language13191261GPT-4.1 Mini
LMArena Text: entertainment and media13001243GPT-4.1 Mini
LMArena Vision: English11911101GPT-4.1 Mini
LMArena Vision: Chinese1186Not measuredOnly GPT-4.1 Mini measured
LMArena Vision: reading text1188Not measuredOnly GPT-4.1 Mini measured
LMArena Vision: humor1182Not measuredOnly GPT-4.1 Mini measured
LMArena Vision: captioning1174Not measuredOnly GPT-4.1 Mini measured
LMArena Vision: recognizing things1187Not measuredOnly GPT-4.1 Mini measured
LMArena Vision: creative writing1182Not measuredOnly GPT-4.1 Mini measured
Neither model has a result on the other 16 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GPT-4.1 Mini and GPT-4.1 Nano
BenchmarkGPT-4.1 MiniGPT-4.1 NanoLeader
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-4.1 Mini$0.0012Fits in one request
GPT-4.1 Nano$0.0003Fits in one request

GPT-4.1 Nano costs less

Repository review

50K input + 3K output tokens

GPT-4.1 Mini$0.025Fits in one request
GPT-4.1 Nano$0.0062Fits in one request

GPT-4.1 Nano costs less

Agent loop

200K input + 10K output tokens

GPT-4.1 Mini$0.096Fits in one request
GPT-4.1 Nano$0.024Fits in one request

GPT-4.1 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-4.1 Mini and GPT-4.1 Nano
SpecificationOpenAI GPT-4.1 MiniOpenAI GPT-4.1 Nano
MakerOpenAIOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedApril 14, 2025April 14, 2025
Context window1.05M1.05M
Input $/Mtok$0.40$0.10
Output $/Mtok$1.60$0.40
Coding index28.58.9
Independent measurements behind it21
Cost per solved task$1.818Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-4.1 Mini better than GPT-4.1 Nano for coding?

Pick GPT-4.1 Mini for coding: it scores 28.5 on the Vibe Coding Index against 8.9 for GPT-4.1 Nano. Pick GPT-4.1 Nano when output tokens cost 4.0x less, or when the benchmark that matters is Design Arena: 3D, where it leads.

Which is cheaper, GPT-4.1 Mini or GPT-4.1 Nano?

GPT-4.1 Mini costs $0.40 per million input tokens and $1.60 per million output tokens; GPT-4.1 Nano costs $0.10 and $0.40. Cost per solved task is known for GPT-4.1 Mini only.

Which has the bigger context window, GPT-4.1 Mini or GPT-4.1 Nano?

Neither: both take 1.05M tokens.

How many benchmarks is the GPT-4.1 Mini vs GPT-4.1 Nano comparison based on?

69 benchmarks have a result for both models as of October 21, 2026. GPT-4.1 Mini's coding index rests on 2 independent benchmarks and GPT-4.1 Nano's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next