‹ PublicAI Index
The LLM benchmark aggregator.
LongWriter Glm4 9B
Z.ai · 9B
Strongest in Jailbreak resistance (#118 of 272), weakest in Toxicity avoidance (#223 of 272). Above par in 4 of 6 scopes. Among the models it meets almost everywhere, it finishes behind Claude Sonnet 5 and Claude Opus 5.5 and ahead of DeepSeek Chat and Gemma 4 26B A4B.
§ 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.
Safety50.6−10.9#165/3361/3
Jailbreak resistance55.1−9.8#118/2721/1
Harm refusal51.7−9.9#131/2991/2
Secure code51.2−15.3#149/2741/1
Fairness47.2−22.8#154/2991/2
Toxicity avoidance46.7−12.3#223/2721/1
§ 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 LongWriter Glm4 9B, 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 8.5%Enkrypt · Harmful content risk 42.2%Enkrypt · CBRN risk 5.7%Enkrypt · Toxicity risk 8.6%Enkrypt · Bias risk 82.2%Enkrypt · Insecure code risk 31.1%
Badge
[](https://publicai.io/model-index/m/longwriter-glm4-9b)