US (Other)
trinity-large-thinking
Public AI IQ model profile with source-backed benchmark results and derived ranking context. Open Source
IQ
91
EQ
n/a
Effective Cost
n/a
IQ Rank
#73
IQ Dimensions
| Dimension | IQ | Coverage |
|---|---|---|
| mathematical reasoning | 95 | 0/5 |
| scientific reasoning | 119 | 4/4 |
| abstract reasoning | 71 | 0/3 |
| app building | 77 | 1/4 |
| production engineering | 94 | 0/7 |
| computer use | 96 | 1/7 |
| reliability | 88 | 3/3 |
Benchmark Results
| Benchmark | Score | Dimension |
|---|---|---|
| AA Omniscience | -44 | reliability |
| Arena.ai Overall | 1369 | |
| BullshitBench v2 | 13 | reliability |
| CritPt | 1 | scientific-reasoning |
| GPQA Diamond | 75 | scientific-reasoning |
| Humanity's Last Exam | 15 | scientific-reasoning |
| IFBench | 56 | reliability |
| SciCode | 36 | scientific-reasoning |
| Terminal-Bench Hard | 23 | computer-use |
| Arena.ai WebDev | 1245 | app-building |