GPT-5.4 Mini vs Kimi K2.5
Head to head on 25 shared benchmarks, with price, context window, and release dates.
Benchmark scores
42evals · sort by any column, ⤢ to expand fullscreen
| Model | |||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Kimi K2.5 Moonshot AI | 68.3% | 1576elo | 74.9 | 35.8% | 27.9% | 78.9% | 13.2% | 43.7% | 78.5% | 72.5% | 77.9% | 61.4% | 57.4% | 77.7% | 84.9% | 76.0% | 66.5% | 49.9% | 39.6% | 74.2% | 18.9% | 69.4 | 26.3% | 17.5% | 81.3% | 96.1 | - | 39.0% | - | - | 95.4 | 10421 | 62.0% | 39.3% | 76.4% | 87.1 | 63.3% | - | 73 | 67.3% | 70.8% | - | 56.8% |
| GPT-5.4 Mini OpenAI | 60.9% | 1662elo | 73 | 45.4% | 28.3% | 60.6% | 5.9% | 38.8% | 80.3% | 70.9% | 71.6% | 70.9% | 60.3% | 71.5% | 78.6% | 72.5% | 63.5% | 50.8% | 39.6% | 71.2% | 18.2% | 61.7 | 39.6% | 48.0% | 23.4% | - | 37.0% | - | 21.2% | 3.9% | - | - | - | - | - | - | - | 21.0% | - | - | - | 54.2% | 48.4% |
2 / 2 models
| Attribute | GPT-5.4 Mini OpenAI | Kimi K2.5 Moonshot AI |
|---|---|---|
| Description | GPT-5.4 mini is an AI model from OpenAI. | Kimi K2.5 is an AI model from Kimi. |
| Family | GPT | Kimi |
| Params | - | - |
| Open weights | closed | closed |
| License | proprietary | Modified MIT |
| Released | 17 Mar 2026 | 27 Jan 2026 |
| Context | 400,000 | 262,144 |
| Input $/1M | $0.75 | $0.60 |
| Output $/1M | $4.50 | $3.00 |
| Cache read $/1M | $0.075 | $0.100 |
| Cache write $/1M | - | - |
| Speed | 0 tok/s | 0 tok/s |
| Latency | 0.00 s | 0.00 s |
| Modalities | file, image, text | text, image |
| Providers | - | - |
| Publisher | OpenAI | Moonshot AI |
| Reported scores | GPQA Diamond78.9% (AA) Humanity's Last Exam (HLE)13.2% (AA) IFBench43.7% (AA) SciCode39.6% (AA) τ²-bench (Tau²-bench)81.3% (AA) |
FAQ
- Is GPT-5.4 Mini or Kimi K2.5 better?
- Across the 25 benchmarks both models report on Sophon, Kimi K2.5 leads on 15 of them. Which one is "better" depends on the benchmark - the table above shows every shared score.
- How do GPT-5.4 Mini and Kimi K2.5 compare on GPQA Diamond?
- GPT-5.4 Mini scores 60.6% and Kimi K2.5 scores 78.9% on GPQA Diamond.
- Which is cheaper, GPT-5.4 Mini or Kimi K2.5?
- Kimi K2.5 is cheaper at $3.00 per million output tokens against $4.50 for GPT-5.4 Mini - about 1.5x.
- When were GPT-5.4 Mini and Kimi K2.5 released?
- GPT-5.4 Mini was released 17 Mar 2026 by OpenAI. Kimi K2.5 was released 27 Jan 2026 by Moonshot AI.
Comparing something else? Build your own side-by-side.