Compare / Kimi K3 vs GPT-5.4
Kimi K3 vs GPT-5.4
23 results from 6 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Kimi K3 | GPT-5.4 |
|---|---|---|---|
| APEX-Agentsreported | Tasks passed% of tasks | 50.6% | 52.4% |
| Arena (formerly LMArena)reported | Confirmed task successIPS effect estimate | 0.09 | -0.04 |
| Arena (formerly LMArena)reported | Praise over complaintIPS effect estimate | 0.07 | -0.01 |
| Arena (formerly LMArena)reported | SteerabilityIPS effect estimate | -0.01 | 0.02 |
| Arena (formerly LMArena)reported | Tool groundingIPS effect estimate | 0.00 | 0.00 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,472 | 1,485 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,489 | 1,482 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,458 | 1,443 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,533 | 1,515 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,486 | 1,471 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,488 | 1,475 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,475 | 1,465 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 76.2 | 71.1 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 43.6 | 39.0 |
| Artificial Analysisreported | GPQA Diamond% of questions | 93.5% | 92.0% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 46.9% | 43.7% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 88.7% | 82.0% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 85.0% | 78.3% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 46.0% | 39.6% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 50.6% | 45.1% |
| tau2-benchreported | Consistency (pass^4)% of tasks | 17.5% | 21.6% |
| tau2-benchreported | Task success (pass^1)% of tasks | 37.1% | 39.4% |
| Vending-Bench 2reported | Money after a yearUS dollars | $5,165.03 | $6,144.18 |
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.