Compare / Phi 4 vs Kimi K2 0905
Phi 4 vs Kimi K2 0905
11 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Phi 4 | Kimi K2 0905 |
|---|---|---|---|
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,256 | 1,404 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,210 | 1,382 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,268 | 1,417 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,245 | 1,390 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,256 | 1,419 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,229 | 1,382 |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 19.0% | 22.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.34 | 0.31 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 25.7 | 54.4 |
| Vectara Hallucination Leaderboardreported | Answer rate% of documents | 80.7% | 98.6% |
| Vectara Hallucination Leaderboardreported | Hallucination rate% of summaries | 3.7% | 17.9% |
Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.