Compare / Kimi K2.6 vs GPT-5.4
Kimi K2.6 vs GPT-5.4
25 results from 5 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Kimi K2.6 | GPT-5.4 |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,450 | 1,470 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,456 | 1,485 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,455 | 1,482 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,433 | 1,443 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,503 | 1,515 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,456 | 1,471 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,461 | 1,475 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,441 | 1,465 |
| Microsoft STATE-Benchreported | Consistency (pass^5)% of tasks | 29.3% | 38.0% |
| Microsoft STATE-Benchreported | Customer support task success (pass@1)% of tasks | 45.1% | 57.6% |
| Microsoft STATE-Benchreported | Shopping assistant task success (pass@1)% of tasks | 47.9% | 53.6% |
| Microsoft STATE-Benchreported | Task success (pass@1)% of tasks | 48.3% | 55.7% |
| Microsoft STATE-Benchreported | Travel task success (pass@1)% of tasks | 51.9% | 55.9% |
| Microsoft STATE-Benchreported | User experiencescore from 1 to 5 | 3.37 | 3.49 |
| OpenHands Indexreported | Average% resolved | 57.1% | 64.3% |
| OpenHands Indexreported | Front end (SWE-Bench Multimodal)% resolved | 41.2% | 36.8% |
| OpenHands Indexreported | Greenfield (Commit0)% resolved | 25.0% | 56.2% |
| OpenHands Indexreported | Information gathering (GAIA)% resolved | 74.5% | 82.4% |
| OpenHands Indexreported | Issue resolution (SWE-Bench)% resolved | 74.6% | 75.6% |
| OpenHands Indexreported | Testing (SWT-Bench)% resolved | 70.4% | 70.4% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 18.0% | 30.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.35 | 0.40 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 64.4 | 68.8 |
| Vectara Hallucination Leaderboardreported | Answer rate% of documents | 99.7% | 99.9% |
| Vectara Hallucination Leaderboardreported | Hallucination rate% of summaries | 10.8% | 7.0% |
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.