‹ PublicAI Index
The LLM benchmark aggregator.
GPT-5.2
OpenAI
Strongest in Secure code (#1 of 270), weakest in Toxicity avoidance (#125 of 268). Above par in 25 of 27 scopes. Among the models it meets almost everywhere, it finishes behind Claude Fable 5 and Claude Fable 5.1 and ahead of GPT-5.6 Terra and Grok 4.5.
§ 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.
Professional57−5.4#16/1941/2
Finance57.7−4.7#21/1621/1
Medical57.9−8.6#23/1491/1
Legal57.2−8.7#25/1731/1
Safety57.7−3.7#19/3292/3
Secure code66.2leads#1/2701/1
Fairness62.5−7.5#28/2921/2
Jailbreak resistance60.5−4.5#36/2681/1
Harm refusal56.8−4.7#37/2921/2
Factual grounding47.8−22.9#68/1041/1
Toxicity avoidance54.2−4.8#125/2681/1
Reasoning56.5−8.9#33/1954/5
Mathematics58.5−5.3#18/1522/2
Science60.9−2.6#20/1221/1
Reasoning51.6−17.4#58/1472/3
Agents55.4−11#53/2981/6
Tool use63.1−10.9#13/811/1
Knowledge52.6−9.9#75/1971/3
Academic knowledge56.3−7.6#45/1231/1
Core abilities51.4−13.9#83/1992/4
General intelligence60.1−9.3#26/1781/3
Language52.3−14.3#30/621/1
Instruction following39−34.8#52/621/1
Human preference60.2−8.6#84/3541/1
Human preference60.2−8.6#84/3541/1
Coding50.2−16#87/1792/6
Agentic coding50.2−16.7#85/1732/5
§ 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 GPT-5.2, left for the other.
§ 3 · Sources
Where the numbers come from
9 publications, 33 figures. Every one links to the page it was read from.
LMArena Text 1437
ARC-AGI-2 72.9%
LiveBench 1.0 74.6LiveBench 1.0 · Reasoning 83.2LiveBench 1.0 · Coding 76.1LiveBench 1.0 · Agentic Coding 50.3LiveBench 1.0 · Mathematics 93.2LiveBench 1.0 · Data Analysis 78.2LiveBench 1.0 · Language 79.8LiveBench 1.0 · Instruction Following 61.8
BFCL v4 55.87%
Kagi LLM Benchmark 73.3%
SimpleBench 45.8%
Vals · CaseLaw 66.02%Vals · LegalBench 82.76%Vals · CorpFin 65.89%Vals · TaxEval 75.76%Vals · MortgageTax 67.13%Vals · MedQA 94.13%Vals · MedCode 49.75%Vals · MedScribe 84.39%Vals · SWE-bench Verified 75.8% (Mini-SWE-agent)Vals · Vibe Code Bench 53.5% (OpenHands)Vals · GPQA Diamond 91.67%Vals · MMLU Pro 86.23%Vals · AIME 96.88%
Enkrypt · Jailbreak risk 4%Enkrypt · Harmful content risk 8.3%Enkrypt · CBRN risk 10.2%Enkrypt · Toxicity risk 3.3%Enkrypt · Bias risk 58.4%Enkrypt · Insecure code risk 0%
Vectara · Factual consistency 89.2%
Badge
[](https://publicai.io/model-index/m/gpt-5-2)