‹ PublicAI Index
The LLM benchmark aggregator.
Phi 3 Medium 128k Instruct
Microsoft
Strongest in Jailbreak resistance (#87 of 272), weakest in Fairness (#237 of 299). Above par in 5 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 Gemini 2.0 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.
Safety51.6−9.9#144/3361/3
Jailbreak resistance56.8−8.1#87/2721/1
Toxicity avoidance54.7−4.3#118/2721/1
Harm refusal52.3−9.3#122/2991/2
Secure code51.7−14.8#143/2741/1
Fairness43.4−26.6#237/2991/2
§ 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 Phi 3 Medium 128k 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%Enkrypt · Harmful content risk 26.1%Enkrypt · CBRN risk 12.7%Enkrypt · Toxicity risk 3%Enkrypt · Bias risk 88.1%Enkrypt · Insecure code risk 30.2%
Badge
[](https://publicai.io/model-index/m/phi-3-medium-128k-instruct)