‹ PublicAI Index
The LLM benchmark aggregator.
Grok 4.20
SpaceXAI
Strongest in General intelligence (#21 of 178, on 1 of its 3 boards), weakest in Coding (#165 of 179). Above par in 10 of 16 scopes. Among the models it meets almost everywhere, it finishes behind Claude Fable 5 and Claude Fable 5.1 and ahead of MiMo V2.5 Pro and Grok 4.7.
§ 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.
Core abilities57.6−7.7#23/1991/4
General intelligence60.9−8.5#21/1781/3
Reasoning57.4−8#27/1952/5
Mathematics58−5.8#23/1521/2
Science58.8−4.7#34/1221/1
Reasoning55.1−13.9#45/1471/3
Human preference63.6−5.2#39/3541/1
Human preference63.6−5.2#39/3541/1
Knowledge52.6−9.9#74/1971/3
Academic knowledge56.3−7.6#43/1231/1
Professional44−18.4#156/1941/2
Medical43.1−23.4#125/1491/1
Finance44.3−18.1#126/1621/1
Legal42.9−23#139/1731/1
Coding40.1−26.1#165/1791/6
Agentic coding39.5−27.4#160/1731/5
§ 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 4.20, left for the other.
§ 3 · Sources
Where the numbers come from
4 publications, 19 figures. Every one links to the page it was read from.
LMArena Text 1472
ARC-AGI-2 65.1%
Kagi LLM Benchmark 75%
Vals · Legal Research Bench 13.94%Vals · CaseLaw 54.45%Vals · LegalBench 77.74%Vals · Finance Agent 28.49%Vals · CorpFin 63.67%Vals · TaxEval 74.12%Vals · MortgageTax 45.35%Vals · MedQA 94.55%Vals · MedCode 32.16%Vals · MedScribe 63.41%Vals · SWE-bench Verified 72.2% (Mini-SWE-agent)Vals · Vibe Code Bench 4.06% (OpenHands)Vals · Code Migration 0.31%Vals · GPQA Diamond 88.64%Vals · MMLU Pro 86.25%Vals · AIME 96.46%
Badge
[](https://publicai.io/model-index/m/grok-4-20)