The Model Index
Up to four models

GPT-5 vs MiniMax-M3

On the index (GPT-4 = 100), GPT-5 scores 149 and MiniMax-M3 147. 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
Presets
GPT-5
149133–164
WeightsClosed
ReleasedAug 2025
Price in / out$1.25 / $10
Context400K
Core tests taken6 of 9
MiniMax-M3
147131–164
WeightsOpen
ReleasedJun 2026
Price in / out$0.30 / $1.2
Context1.05M
Core tests taken4 of 9

GPT-5 and MiniMax-M3 are too close to call (1 point apart, within the uncertainty), ahead on 0 of 1 shared core tests, at 6.5× the price.

Test by testthird-party results

GPT-5MiniMax-M3
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 ProSimpleBenchSimpleBenchMATHSAIME-style maths (OTIS mock)AIME-style maths (OTIS mock)Chess puzzlesChess puzzlesCODINGLiveCodeBenchLiveCodeBenchSWE-benchSWE-benchVibe Code Bench v1.1Vibe Code Bench v1.1AGENTSDTBenchDTBenchLMCALMCAMystery game puzzlesMystery game puzzlesPROFESSIONAL WORKCorpFin v2CorpFin v2LegalBenchLegalBenchMedCodeMedCodeMedScribeMedScribeMortgageTaxMortgageTaxTaxEval v2TaxEval v2
Every result as a table
TestGPT-5MiniMax-M3
Reasoning & knowledge
GPQA Diamond (Epoch AI)86.2%90.9%
GPQA Diamond (Vals AI)85.6%92.7%
MMLU Pro86.5%84.2%
MMMU Pro81.5%81.2%
SimpleBench56.7%45.8%
Maths
AIME-style maths (OTIS mock)91.4%71.1%
Chess puzzles36.0%17.0%
Coding
LiveCodeBench85.9%82.2%
SWE-bench69.0%75.0%
Vibe Code Bench v1.120.1%47.6%
Agents
DTBench90.7%78.9%
LMCA40.0%33.7%
Mystery game puzzles23.0%8.0%
Professional work
CorpFin v261.1%68.1%
LegalBench86.0%85.4%
MedCode49.6%46.3%
MedScribe83.7%87.3%
MortgageTax65.5%68.4%
TaxEval v273.4%72.7%
Task length · METR42 sec

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

This page as a chartDownload PNG