‹ PublicAI Index
The LLM benchmark aggregator.
GPT-4.1
OpenAI
Strongest in Legal (#12 of 173), weakest in Toxicity avoidance (#184 of 268). Above par in 18 of 26 scopes. Among the models it meets almost everywhere, it finishes behind Claude Fable 5 and Claude Fable 5.1 and ahead of GLM-5.3 Flash and DeepSeek V4 Flash.
§ 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.
Professional56.2−6.2#23/1941/2
Legal60.1−5.8#12/1731/1
Finance55.9−6.5#38/1621/1
Medical54.1−12.4#54/1491/1
Safety55.8−5.6#39/3293/3
Safe-prompt compliance58.4−2#12/821/1
Factual grounding61−9.7#21/1041/1
Harm refusal56.2−5.3#44/2922/2
Secure code60.8−5.4#57/2701/1
Jailbreak resistance56.7−8.3#88/2681/1
Fairness50.3−19.7#110/2922/2
Toxicity avoidance51−8#184/2681/1
Agents54.9−11.5#59/2981/6
Tool use61.7−12.3#16/811/1
Coding51.9−14.3#73/1791/6
Agentic coding52−14.9#69/1731/5
Core abilities49.9−15.4#99/1991/4
General intelligence49.8−19.6#89/1781/3
Knowledge49.6−12.9#117/1971/3
Academic knowledge49.1−14.8#82/1231/1
Human preference58.1−10.7#117/3541/1
Human preference58.1−10.7#117/3541/1
Reasoning40.5−24.9#167/1953/5
Science42.6−20.9#94/1221/1
Mathematics42.9−20.9#118/1521/2
Reasoning37.6−31.4#131/1472/3
§ 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-4.1, left for the other.
§ 3 · Sources
Where the numbers come from
10 publications, 27 figures. Every one links to the page it was read from.
LMArena Text 1415
ARC-AGI-2 0.4%
Aider polyglot 52.4%
BFCL v4 53.96%
Kagi LLM Benchmark 52.3%
SimpleBench 27%
Vals · CaseLaw 69.88%Vals · LegalBench 83.1%Vals · CorpFin 63.05%Vals · TaxEval 75.06%Vals · MortgageTax 65.94%Vals · MedQA 91.18%Vals · GPQA Diamond 65.4%Vals · MMLU Pro 80.5%Vals · AIME 39.58%
HELM Safety · HarmBench 91.7%HELM Safety · SimpleSafetyTests 100%HELM Safety · Anthropic Red Team 99.3%HELM Safety · BBQ 92.6%HELM Safety · XSTest 97.9%
Enkrypt · Jailbreak risk 7.2%Enkrypt · Harmful content risk 25%Enkrypt · CBRN risk 10%Enkrypt · Toxicity risk 5.5%Enkrypt · Bias risk 82.2%Enkrypt · Insecure code risk 11.1%
Vectara · Factual consistency 94.4%
Badge
[](https://publicai.io/model-index/m/gpt-4-1)