‹ PublicAI Index
The LLM benchmark aggregator.
MiniMax M2
MiniMax
Strongest in Safety (#13 of 329, on 1 of its 3 boards), weakest in Human preference (#181 of 354). Above par in 10 of 12 scopes. Among the models it meets almost everywhere, it finishes behind Claude Opus 5.5 and Grok 4.7 and ahead of Gemini 3.6 Flash and GPT-6 Sol.
§ 1 · Profile
What it is good at
Bars run from 50 — the average of the models each source lists — so right of the line is above par. The middle column is the gap to whoever leads that scope.
Safety58.5−2.9#13/3291/3
Fairness69.2−0.8#13/2921/2
Toxicity avoidance57.7−1.3#36/2681/1
Jailbreak resistance58.2−6.8#64/2681/1
Secure code59−7.2#71/2701/1
Harm refusal55−6.5#74/2921/2
Core abilities51.7−13.6#78/1991/4
General intelligence52.5−16.9#66/1781/3
Reasoning42.7−22.7#157/1951/5
Reasoning38.3−30.7#127/1471/3
Human preference51.3−17.5#181/3541/1
Human preference51.3−17.5#181/3541/1
§ 2 · Head to head
What it beats, and what beats it
The same models turn up scope after scope. Counted once: where both were placed, who finished higher. Bars run right for MiniMax M2, left for the other.
§ 3 · Sources
Where the numbers come from
4 publications, 9 figures. Every one links to the page it was read from.
LMArena Text 1344
Kagi LLM Benchmark 57.8%
SimpleBench 25%
Enkrypt · Jailbreak risk 5.9%Enkrypt · Harmful content risk 10.6%Enkrypt · CBRN risk 14%Enkrypt · Toxicity risk 0.9%Enkrypt · Bias risk 48.1%Enkrypt · Insecure code risk 14.7%
Badge
[](https://publicai.io/model-index/m/minimax-m2)