‹ PublicAI Index
The LLM benchmark aggregator.
Grok 2
SpaceXAI
Strongest in Medical (#45 of 149), weakest in Human preference (#191 of 354). Above par in 3 of 11 scopes. Among the models it meets almost everywhere, it finishes behind Claude Fable 5 and Claude Opus 4.6 and ahead of Mistral Medium 3.1 and GPT-4o.
§ 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.
Professional48.9−13.5#113/1941/2
Medical54.7−11.8#45/1491/1
Finance44.8−17.6#124/1621/1
Knowledge47−15.5#141/1971/3
Academic knowledge42.8−21.1#100/1231/1
Reasoning35.7−29.7#186/1952/5
Science32.4−31.1#106/1221/1
Reasoning37.4−31.6#133/1471/3
Mathematics36.4−27.4#134/1521/2
Human preference50.5−18.3#191/3541/1
Human preference50.5−18.3#191/3541/1
§ 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 Grok 2, left for the other.
§ 3 · Sources
Where the numbers come from
3 publications, 8 figures. Every one links to the page it was read from.
LMArena Text 1336
SimpleBench 22.7%
Vals · CorpFin 51.13%Vals · TaxEval 67.05%Vals · MedQA 92.32%Vals · GPQA Diamond 50.76%Vals · MMLU Pro 75.47%Vals · AIME 15.21%
Badge
[](https://publicai.io/model-index/m/grok-2)