‹ PublicAI Index
The LLM benchmark aggregator.
Smollm2 1.7B Instruct
Hugging Face · 1.7B
Strongest in Jailbreak resistance (#191 of 272), weakest in Safety (#324 of 337). Above par in 0 of 8 scopes. Among the models it meets almost everywhere, it finishes behind Claude Opus 5.5 and Claude Opus 4.7 and ahead of Gemma 2B It and PowerLM 3B.
§ 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.
Human preference29.6−37.9#322/3421/1
Human preference29.6−37.9#322/3421/1
Safety40.5−21#324/3371/3
Jailbreak resistance48.2−16.7#191/2721/1
Harm refusal45−16.6#245/3001/2
Toxicity avoidance34.5−24.5#250/2721/1
Secure code26−40.5#272/2741/1
Fairness38.6−31.4#282/3001/2
§ 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 Smollm2 1.7B Instruct, left for the other.
§ 3 · Sources
Where the numbers come from
2 publications, 7 figures. Every one links to the page it was read from.
LMArena Text 1115
Enkrypt · Jailbreak risk 14.3%Enkrypt · Harmful content risk 45.6%Enkrypt · CBRN risk 21.5%Enkrypt · Toxicity risk 17.2%Enkrypt · Bias risk 95.6%Enkrypt · Insecure code risk 94.2%
Badge
[](https://publicai.io/model-index/m/smollm2-1-7b-instruct)