‹ PublicAI Index
The LLM benchmark aggregator.
Qwen2.5 14B Instruct
Alibaba · 14B
Strongest in Toxicity avoidance (#92 of 272), weakest in Fairness (#220 of 299). Above par in 4 of 6 scopes. Among the models it meets almost everywhere, it finishes behind Claude Sonnet 5 and MiniMax M2.7 and ahead of Mistral Large and Sarvam 30B.
§ 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.
Safety51.3−10.2#148/3361/3
Toxicity avoidance55.7−3.3#92/2721/1
Jailbreak resistance55.9−9#102/2721/1
Harm refusal52.1−9.5#124/2991/2
Secure code49.7−16.8#159/2741/1
Fairness44−26#220/2991/2
§ 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 Qwen2.5 14B Instruct, left for the other.
§ 3 · Sources
Where the numbers come from
1 publication, 6 figures. Every one links to the page it was read from.
Enkrypt · Jailbreak risk 7.8%Enkrypt · Harmful content risk 30.6%Enkrypt · CBRN risk 10.8%Enkrypt · Toxicity risk 2.3%Enkrypt · Bias risk 87.1%Enkrypt · Insecure code risk 34.2%
Badge
[](https://publicai.io/model-index/m/qwen2-5-14b-instruct)