‹ PublicAI Index
The LLM benchmark aggregator.
Jamba 1.6 Large
AI21
Strongest in Legal (#120 of 173), weakest in Safety (#234 of 329). Above par in 1 of 15 scopes. Among the models it meets almost everywhere, it finishes behind Claude Opus 4.8 and Claude Opus 4.7 and ahead of Jamba 1.5 Large and Mixtral 8x7B Instruct.
§ 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.
Knowledge40−22.5#181/1971/3
Academic knowledge26−37.9#121/1231/1
Professional39.6−22.8#183/1941/2
Legal45.2−20.7#120/1731/1
Medical32.9−33.6#142/1491/1
Finance36.4−26#149/1621/1
Reasoning33.6−31.8#193/1951/5
Science26.1−37.4#120/1221/1
Mathematics32.9−30.9#148/1521/2
Safety47.9−13.5#234/3291/3
Jailbreak resistance52−13#155/2681/1
Fairness46.2−23.8#160/2921/2
Harm refusal47.7−13.8#191/2921/2
Secure code44.2−22#197/2701/1
Toxicity avoidance48.9−10.1#205/2681/1
§ 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 Jamba 1.6 Large, left for the other.
§ 3 · Sources
Where the numbers come from
2 publications, 13 figures. Every one links to the page it was read from.
Vals · LegalBench 71.96%Vals · CorpFin 41.53%Vals · TaxEval 60.88%Vals · MedQA 50.7%Vals · GPQA Diamond 36.11%Vals · MMLU Pro 49.78%Vals · AIME 0.42%
Enkrypt · Jailbreak risk 11.1%Enkrypt · Harmful content risk 50.6%Enkrypt · CBRN risk 11.7%Enkrypt · Toxicity risk 6.9%Enkrypt · Bias risk 83.7%Enkrypt · Insecure code risk 44.9%
Badge
[](https://publicai.io/model-index/m/jamba-1-6-large)