Kimi
kimi-k2.6
Public AI IQ model profile with source-backed benchmark results and derived ranking context. Open Source
IQ
117
EQ
116
Effective Cost
$5.0815
IQ Rank
#10
IQ Dimensions
| Dimension | IQ | Coverage |
|---|---|---|
| mathematical reasoning | 125 | 4/5 |
| scientific reasoning | 135 | 4/4 |
| abstract reasoning | 82 | 0/3 |
| app building | 127 | 4/4 |
| production engineering | 114 | 6/7 |
| computer use | 125 | 6/7 |
| reliability | 108 | 3/3 |
Benchmark Results
| Benchmark | Score | Dimension |
|---|---|---|
| AA Omniscience | 6 | reliability |
| Agents' Last Exam | 9.2 | computer-use |
| Arena.ai Overall | 1460 | |
| BrowseComp | 86.3 | computer-use |
| BullshitBench v2 | 50 | reliability |
| CritPt | 8 | scientific-reasoning |
| DeepSWE | 23.89 | production-engineering |
| DesignArena Frontend | 1275 | app-building |
| DesignArena Full Stack | 1240 | app-building |
| EQ-Bench 3 Elo | 1575.7 | |
| FrontierCode Diamond | 3.8 | production-engineering |
| FrontierMath Tier 1-3 | 57.2 | mathematical-reasoning |