‹ PublicAI Index
The LLM benchmark aggregator.
Trinity Large
Arcee · 399B · open weights
Strongest in Factual grounding (#29 of 101), weakest in Agents (#228 of 268). Above par in 4 of 11 scopes. Among the models it meets almost everywhere, it finishes behind Claude Opus 5.5 and Muse Spark 1.1 and ahead of DeepSeek R1 and Mistral Small.
§ 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.
Professional48.4−13.2#97/1681/1
Legal48.8−18.3#86/1511/1
Finance46.7−16.2#104/1511/1
Safety52.4−9.1#118/3371/3
Factual grounding57.8−12.8#29/1011/1
Human preference53.9−13.6#150/3421/1
Human preference53.9−13.6#150/3421/1
Core abilities43.8−23.6#162/2041/3
General intelligence41−29.1#166/2041/3
Agents41.3−26.8#228/2682/5
Knowledge work38.7−34.9#139/1782/2
§ 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 Trinity Large, left for the other.
§ 3 · Sources
Where the numbers come from
6 publications, 7 figures. Every one links to the page it was read from.
LMArena Text 1368
Artificial Analysis Intelligence Index 11
GDPval-AA 324
AA-Briefcase 255
Vals · CaseLaw 57.88%Vals · CorpFin 54.66%
Vectara · Factual consistency 93.1%
Badge
[](https://publicai.io/model-index/m/trinity-large)