AI BENCHY Compare
MoonshotAI: Kimi K2.5 vs OpenAI: GPT-5.4
Last updated at: 2026-04-16
| Metric | Kimi K2.5 Kimi K2.5 medium | GPT-5.4 GPT-5.4 none |
|---|---|---|
| Score | 7.0 | 5.9 |
| Rank | #45 | #65 |
| Consistency | 6.8 | 9.1 |
| Tests Correct | ||
| Attempt pass rate | 72.2% | 42.6% |
| Flaky tests | 7 | 2 |
| Total Runs | 54 | 54 |
| Cost per result | 2.444 | 1.477 |
| Total Cost | $0.220 | $0.104 |
| Input Price | $0.383 / 1M | $2.500 / 1M |
| Output Price | $1.720 / 1M | $15.000 / 1M |
| Output Tokens | 42,176 | 2,317 |
| Reasoning Tokens | 84,870 | 0 |
| Response Time (avg) | 72.43s | 1.51s |
| Response Time (max) | 150.77s | 2.95s |
| Response Time (total) | 796.70s | 27.21s |
Score vs Total Cost
Response Time (avg)
Score vs Response Time (avg)
Total Output Tokens
Score vs Total Output Tokens
Category Breakdown
Quick Compare
Switch Comparison Pair
Gemma 4 31BnoneFree AvailablevsKimi K2.5mediumMiniMax M2.5mediumFree AvailablevsGPT-5.4noneKimi K2.5mediumvsQwen3.5 Plus 2026-02-15noneMistral Small 4mediumvsGPT-5.4noneKimi K2.5mediumvsGLM 5noneClaude Sonnet 4.6nonevsKimi K2.5mediumKimi K2.5mediumvsMiMo-V2-OmninoneMercury 2mediumvsGPT-5.4noneMiniMax M2.7mediumvsGPT-5.4noneGPT-5.4nonevsElephantmediumKimi K2.5mediumvsGPT-5.3 ChatnoneGPT-5.4nonevsGrok 4.1 Fastmedium