Neura Intelligence Index

AI Model Benchmarks

DeepSeek-V3.1 vs MiniMax M2 — Benchmark Comparison | Neura Market

Back to Leaderboard

DeepSeek-V3.1 vs MiniMax M2

DeepSeek 🇨🇳

45.0

#173—

MiniMax 🇨🇳

45.5

#170—

Category Comparison

Benchmark-by-Benchmark

reasoning

Benchmark	DeepSeek-V3.1	MiniMax M2
GPQA Diamond	74.9	78
Humanity's Last Exam	15.9	12.5

math

Benchmark	DeepSeek-V3.1	MiniMax M2
AIME 2025	49.8	78

coding

Benchmark	DeepSeek-V3.1	MiniMax M2
SWE-bench Verified	66	69.4
TerminalBench	31.3	46.3
SciCode	—	36

general

Benchmark	DeepSeek-V3.1	MiniMax M2
SimpleQA	93.4	—

agent

Benchmark	DeepSeek-V3.1	MiniMax M2
BrowseComp	30	44

Category Winners

coding

MiniMax M2

53.1

math

MiniMax M2

47.9

reasoning

MiniMax M2

43.5

general

DeepSeek-V3.1

98.3

agent

MiniMax M2

15.5