The Model Index
Up to four models

o3 vs Llama 4 Maverick

On the index (GPT-4 = 100), o3 scores 147 and Llama 4 Maverick 116. 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 o3
Presets
o3
147131–163
WeightsClosed
ReleasedApr 2025
Price in / out$2.0 / $8.0
Context200K
Core tests taken6 of 9
Llama 4 Maverick
116100–132
WeightsOpen
ReleasedApr 2025
Price in / out$0.35 / $1.0
Context524K
Core tests taken3 of 9

o3 leads Llama 4 Maverick by 31 points, ahead on 3 of 3 shared core tests, at 6.8× the price.

Test by testthird-party results

o3Llama 4 Maverick
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)Humanity's Last ExamHumanity's Last ExamGPQA Diamond (Vals AI)GPQA Diamond (Vals AI)MMLU ProMMLU ProMMMU ProMMMU ProFiction.LiveBenchFiction.LiveBenchGeoBenchGeoBenchSimpleBenchSimpleBenchMATHSAIME-style maths (OTIS mock)AIME-style maths (OTIS mock)MATH (level 5)MATH (level 5)MATH 500MATH 500MGSMMGSMFrontierMath (2025 set)FrontierMath (2025 set)CODINGAider PolyglotAider PolyglotLiveCodeBenchLiveCodeBenchWeirdMLWeirdMLAGENTSDTBenchDTBenchLMCALMCAPROFESSIONAL WORKCorpFin v2CorpFin v2LegalBenchLegalBenchMedCodeMedCodeMedScribeMedScribeMortgageTaxMortgageTaxTaxEval v2TaxEval v2NOVEL PROBLEMSARC-AGI-2ARC-AGI-2ARC-AGI-1ARC-AGI-1
Every result as a table
Testo3Llama 4 Maverick
Reasoning & knowledge
GPQA Diamond (Epoch AI)81.8%67.0%
Humanity's Last Exam20.3%5.7%
GPQA Diamond (Vals AI)84.1%69.4%
MMLU Pro85.6%79.4%
MMMU Pro80.4%71.7%
Fiction.LiveBench88.9%46.2%
GeoBench74.0%52.0%
SimpleBench53.1%27.7%
Maths
AIME-style maths (OTIS mock)84.4%20.6%
MATH (level 5)97.8%73.0%
MATH 50094.6%85.2%
MGSM91.8%92.4%
FrontierMath (2025 set)18.7%0.7%
Coding
Aider Polyglot81.3%15.6%
LiveCodeBench83.9%47.3%
WeirdML52.4%24.5%
Agents
DTBench84.8%61.9%
LMCA39.7%15.9%
Professional work
CorpFin v259.7%49.7%
LegalBench83.8%77.8%
MedCode47.3%36.5%
MedScribe76.6%54.2%
MortgageTax65.7%58.5%
TaxEval v274.6%66.6%
Novel problems
ARC-AGI-26.5%0.0%
ARC-AGI-175.7%4.4%
Task length · METR39 sec

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

This page as a chartDownload PNG