The Model Index
Up to four models

GPT-5 nano vs GPT-4.1 nano

On the index (GPT-4 = 100), GPT-5 nano scores 125 and GPT-4.1 nano 109. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.

Next to GPT-5 nano
Presets
GPT-5 nano
125109–141
WeightsClosed
ReleasedAug 2025
Price in / out$0.050 / $0.40
Context400K
Core tests taken3 of 9
GPT-4.1 nano
10993–125
WeightsClosed
ReleasedApr 2025
Price in / out$0.10 / $0.40
Context1.05M
Core tests taken2 of 9

GPT-5 nano leads GPT-4.1 nano by 16 points, ahead on 2 of 2 shared core tests, at a similar price.

Test by testthird-party results

GPT-5 nanoGPT-4.1 nano
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)GPQA Diamond (Vals AI)GPQA Diamond (Vals AI)MMLU ProMMLU ProMMMU ProMMMU ProFiction.LiveBenchFiction.LiveBenchSimpleQA VerifiedSimpleQA VerifiedMATHSAIME-style maths (OTIS mock)AIME-style maths (OTIS mock)MATH (level 5)MATH (level 5)MATH 500MATH 500MGSMMGSMChess puzzlesChess puzzlesFrontierMath (2025 set)FrontierMath (2025 set)CODINGLiveCodeBenchLiveCodeBenchWeirdMLWeirdMLAGENTSDTBenchDTBenchLMCALMCAPROFESSIONAL WORKLegalBenchLegalBenchMortgageTaxMortgageTaxTaxEval v2TaxEval v2NOVEL PROBLEMSARC-AGI-2ARC-AGI-2ARC-AGI-1ARC-AGI-1
Every result as a table
TestGPT-5 nanoGPT-4.1 nano
Reasoning & knowledge
GPQA Diamond (Epoch AI)69.4%48.9%
GPQA Diamond (Vals AI)63.4%50.8%
MMLU Pro76.1%63.5%
MMMU Pro70.9%55.0%
Fiction.LiveBench44.4%25.0%
SimpleQA Verified11.7%6.0%
Maths
AIME-style maths (OTIS mock)81.1%28.9%
MATH (level 5)95.2%70.0%
MATH 50093.8%80.2%
MGSM89.3%69.3%
Chess puzzles22.0%0.0%
FrontierMath (2025 set)8.3%1.0%
Coding
LiveCodeBench70.2%42.7%
WeirdML38.1%19.0%
Agents
DTBench62.7%52.5%
LMCA7.9%5.5%
Professional work
LegalBench50.1%61.1%
MortgageTax53.6%52.8%
TaxEval v267.4%60.8%
Novel problems
ARC-AGI-22.6%0.0%
ARC-AGI-120.7%0.0%

Tests at least two of these models have taken. Bold rows are the core tests.

This page as a chartDownload PNG