Neura Intelligence Index

AI Model Benchmarks

GPT-4o vs Mistral Large 3 (675B Base) — Benchmark Comparison | Neura Market

Back to Leaderboard

GPT-4o vs Mistral Large 3 (675B Base)

OpenAI 🇺🇸

27.1

#233—

Mistral Large 3 (675B Base)

Mistral AI 🇫🇷

26.1

#236—

Category Comparison

Benchmark-by-Benchmark

reasoning

Benchmark	GPT-4o	Mistral Large 3 (675B Base)
GPQA Diamond	70.1	43.9
Humanity's Last Exam	5.3	—

coding

Benchmark	GPT-4o	Mistral Large 3 (675B Base)
SWE-bench Verified	33.2	—

multimodal

Benchmark	GPT-4o	Mistral Large 3 (675B Base)
MMMU	72.2	—
MMMU-Pro	59.9	—
CharXiv Reasoning	58.8	—

general

Benchmark	GPT-4o	Mistral Large 3 (675B Base)
SimpleQA	38.2	23.8
MMMLU	81.4	85.5

agent

Benchmark	GPT-4o	Mistral Large 3 (675B Base)
TAU-Bench Retail	60.3	—

Category Winners

coding

GPT-4o

2.8

reasoning

GPT-4o

32.3

general

GPT-4o

46.2

multimodal

GPT-4o

36.0

agent

GPT-4o

28.4