LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

GPT-4.1 vs GPT-5 Nano

Updated October 21, 2026. 77 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 Replace GPT-5 Nano
GPT-4.1OpenAI
Coding
46.5
Overall
12.7
Output $/M
$8.00
52
GPT-5 NanoOpenAI
Coding
34.8
Overall
13.0
Output $/M
$0.40
25
vs boards won of 77 shared

Decision reading

Pick GPT-4.1 for coding: it scores 46.5 on the Vibe Coding Index against 34.8 for GPT-5 Nano.

Pick GPT-5 Nano when output tokens cost 20.0x less, or when a solved task costs $0.109 against $0.37.

Model AGPT-4.1OpenAI46.5/1002 independent benchmarks$0.37 per solved taskModel BGPT-5 NanoOpenAI34.8/1001 independent benchmark$0.109 per solved task
Shared results
77
GPT-4.1 only
10
GPT-5 Nano only
8
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 46.5 from 2 benchmarks, GPT-5 Nano 34.8 from 1 benchmark.
Widest category
Math, 23.1 points apart on average over 10 shared benchmarks.
Widest single gap
AA Math Index: GPT-4.1 34.7 against GPT-5 Nano 83.7, 49.0 points apart on a 0 to 100 scale.
Output price
GPT-4.1 $8.00 against GPT-5 Nano $0.40 per million output tokens, 20.0x apart.
Context window
GPT-4.1 takes 1.05M tokens, GPT-5 Nano 400K.

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-4.1BarsGPT-5 NanoTies
Coding43
Intelligence02
Agentic work10
Reasoning76
Math28
Knowledge111
Web design24
General251
All shared boards5225
CodingLike-for-likeFour of these boards feed the Vibe Coding Index43
Coding benchmarks for GPT-4.1 and GPT-5 Nano
BenchmarkGPT-4.1GPT-5 NanoLeader
Aider polyglot52.4%Not measuredOnly GPT-4.1 measured
SWE-bench Verified39.6%34.8%GPT-4.1
SWE-bench Multimodal31.1%Not measuredOnly GPT-4.1 measured
LiveCodeBench (Vals AI)54.7%70.2%GPT-5 Nano
LiveCodeBench (Artificial Analysis run)45.7%78.9%GPT-5 Nano
Terminal-Bench Hard (Artificial Analysis run)13.6%17.4%GPT-5 Nano
Terminal-BenchNot measured21.8%Only GPT-5 Nano measured
WeirdML39.0%38.1%GPT-4.1
ALE-Bench558.1Not measuredOnly GPT-4.1 measured
LMArena Text: coding13901351GPT-4.1
LMArena Text: software and IT13911345GPT-4.1
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-4.1 and GPT-5 Nano
BenchmarkGPT-4.1GPT-5 NanoLeader
AA Intelligence Index12.713.0GPT-5 Nano
Epoch Capabilities Index136.7139.2GPT-5 Nano
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
Agentic work benchmarks for GPT-4.1 and GPT-5 Nano
BenchmarkGPT-4.1GPT-5 NanoLeader
τ²-Bench Telecom47.1%36.5%GPT-4.1
Neither model has a result on the other 38 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index76
Reasoning benchmarks for GPT-4.1 and GPT-5 Nano
BenchmarkGPT-4.1GPT-5 NanoLeader
GPQA Diamond65.4%63.4%GPT-4.1
MMMU Pro (Vals AI)72.4%70.9%GPT-4.1
GPQA Diamond (Artificial Analysis run)66.6%67.6%GPT-5 Nano
Humanity's Last Exam (Artificial Analysis run)4.2%9.5%GPT-5 Nano
GPQA Diamond (Epoch AI run)66.9%69.4%GPT-5 Nano
ARC-AGI5.5%20.7%GPT-5 Nano
ARC-AGI-20.4%2.6%GPT-5 Nano
SimpleBench27.0%Not measuredOnly GPT-4.1 measured
Chess puzzles3.0%22.0%GPT-5 Nano
Mystery game puzzlesNot measured9.0%Only GPT-5 Nano measured
VPCTNot measured37.2%Only GPT-5 Nano measured
LMArena Text: hard prompts13841328GPT-4.1
LMArena Text: expert13641321GPT-4.1
LMArena Text: hard prompts in English13931338GPT-4.1
LMArena Vision: diagrams12151141GPT-4.1
LMArena Vision: homework12481157GPT-4.1
Neither model has a result on the other 5 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index28
Math benchmarks for GPT-4.1 and GPT-5 Nano
BenchmarkGPT-4.1GPT-5 NanoLeader
AIME39.6%81.2%GPT-5 Nano
MATH 500 (Vals AI)87.2%93.8%GPT-5 Nano
MGSM (Vals AI)87.7%89.3%GPT-5 Nano
AA Math Index34.783.7GPT-5 Nano
AIME 2025 (Artificial Analysis run)34.7%83.7%GPT-5 Nano
OTIS Mock AIME38.3%81.1%GPT-5 Nano
FrontierMath, tiers 1 to 36.0%20.0%GPT-5 Nano
FrontierMath, tier 4Not measured2.4%Only GPT-5 Nano measured
ProofBenchNot measured12.0%Only GPT-5 Nano measured
MATH level 583.0%95.2%GPT-5 Nano
LMArena Text: math13691317GPT-4.1
LMArena Text: mathematical work13601326GPT-4.1
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 GPT-4.1 and GPT-5 Nano
BenchmarkGPT-4.1GPT-5 NanoLeader
MMLU Pro80.5%76.1%GPT-4.1
LegalBench (Vals AI)83.1%50.1%GPT-4.1
CaseLaw (Vals AI)69.9%52.6%GPT-4.1
CorpFin (Vals AI)63.0%Not measuredOnly GPT-4.1 measured
TaxEval (Vals AI)75.1%67.4%GPT-4.1
MortgageTax (Vals AI)65.9%53.6%GPT-4.1
MedQA (Vals AI)91.2%93.3%GPT-5 Nano
MedCode (Vals AI)Not measured30.4%Only GPT-5 Nano measured
MMLU Pro (Artificial Analysis run)80.6%78.0%GPT-4.1
SimpleQA Verified31.1%11.7%GPT-4.1
GeoBench72.0%Not measuredOnly GPT-4.1 measured
LMArena Text: science13821318GPT-4.1
LMArena Text: medicine and healthcare13791335GPT-4.1
LMArena Text: legal and government13991298GPT-4.1
LMArena Text: business and finance13691311GPT-4.1
Neither model has a result on the other 2 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index24
Web design benchmarks for GPT-4.1 and GPT-5 Nano
BenchmarkGPT-4.1GPT-5 NanoLeader
Design Arena: websites10391101GPT-5 Nano
Design Arena: UI components10001066GPT-5 Nano
Design Arena: game dev10801045GPT-4.1
Design Arena: data viz11041056GPT-4.1
Design Arena: 3D850966GPT-5 Nano
Design Arena: all code categories10261084GPT-5 Nano
Neither model has a result on the other 3 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index251
General benchmarks for GPT-4.1 and GPT-5 Nano
BenchmarkGPT-4.1GPT-5 NanoLeader
SAGE (Vals AI)Not measured30.4%Only GPT-5 Nano measured
MedScribe (Vals AI)Not measured72.9%Only GPT-5 Nano measured
IFBench43.0%67.6%GPT-5 Nano
Long context reasoning (AA-LCR)68.3%45.0%GPT-4.1
DTBench68.3%62.7%GPT-4.1
LMCA25.6%7.9%GPT-4.1
Fiction.LiveBench63.9%44.4%GPT-4.1
LMArena Text13831320GPT-4.1
LMArena Text: instruction following13671306GPT-4.1
LMArena Text: creative writing13631249GPT-4.1
LMArena Vision12101159GPT-4.1
LMArena Text: English13921321GPT-4.1
LMArena Text: non-English13701313GPT-4.1
LMArena Text: Chinese13821356GPT-4.1
LMArena Text: German13811327GPT-4.1
LMArena Text: Japanese13191227GPT-4.1
LMArena Text: Korean13381269GPT-4.1
LMArena Text: French1382Not measuredOnly GPT-4.1 measured
LMArena Text: Spanish13761360GPT-4.1
LMArena Text: Russian13771296GPT-4.1
LMArena Text: Polish13891314GPT-4.1
LMArena Text: multi-turn13971311GPT-4.1
LMArena Text: longer queries13851312GPT-4.1
LMArena Text: writing and language13711283GPT-4.1
LMArena Text: entertainment and media13651268GPT-4.1
LMArena Vision: English12091170GPT-4.1
LMArena Vision: Chinese12261176GPT-4.1
LMArena Vision: reading text12181152GPT-4.1
LMArena Vision: humor11961110GPT-4.1
LMArena Vision: captioning1207Not measuredOnly GPT-4.1 measured
LMArena Vision: recognizing things1208Not measuredOnly GPT-4.1 measured
LMArena Vision: creative writing1210Not measuredOnly GPT-4.1 measured
Neither model has a result on the other 14 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for GPT-4.1 and GPT-5 Nano
BenchmarkGPT-4.1GPT-5 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$0.0060Fits in one request
GPT-5 Nano$0.0003Fits in one request

GPT-5 Nano costs less

Repository review

50K input + 3K output tokens

GPT-4.1$0.124Fits in one request
GPT-5 Nano$0.0037Fits in one request

GPT-5 Nano costs less

Agent loop

200K input + 10K output tokens

GPT-4.1$0.480Fits in one request
GPT-5 Nano$0.014Fits in one request

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-4.1 and GPT-5 Nano
SpecificationOpenAI GPT-4.1OpenAI GPT-5 Nano
MakerOpenAIOpenAI
WeightsNo public weights listedNo public weights listed
ReleasedApril 14, 2025August 7, 2025
Context window1.05M400K
Input $/Mtok$2.00$0.05
Output $/Mtok$8.00$0.40
Coding index46.534.8
Independent measurements behind it21
Cost per solved task$0.37$0.109
SourceOpenRouter listingOpenRouter listing

Questions

Is GPT-4.1 better than GPT-5 Nano for coding?

Pick GPT-4.1 for coding: it scores 46.5 on the Vibe Coding Index against 34.8 for GPT-5 Nano. Pick GPT-5 Nano when output tokens cost 20.0x less, or when a solved task costs $0.109 against $0.37.

Which is cheaper, GPT-4.1 or GPT-5 Nano?

GPT-4.1 costs $2.00 per million input tokens and $8.00 per million output tokens; GPT-5 Nano costs $0.05 and $0.40. Per solved SWE-bench Verified task, GPT-4.1 costs $0.37 and GPT-5 Nano $0.109.

Which has the bigger context window, GPT-4.1 or GPT-5 Nano?

GPT-4.1, with 1.05M tokens against 400K.

How many benchmarks is the GPT-4.1 vs GPT-5 Nano comparison based on?

77 benchmarks have a result for both models as of October 21, 2026. GPT-4.1's coding index rests on 2 independent benchmarks and GPT-5 Nano's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next