Real large model comparisons, benchmarks and hands-on evaluations by Will
| Model | Chinese | Japanese | Coding | Reasoning | Price |
|---|---|---|---|---|---|
| ❄️Kimi K2.6 | ⭐⭐⭐⭐⭐ | ⭐⭐⭐⭐ | ⭐⭐⭐⭐ | ⭐⭐⭐⭐⭐ | 💰 |
| 🌸Claude Opus 4.7 | ⭐⭐⭐⭐⭐ | ⭐⭐⭐⭐⭐ | ⭐⭐⭐⭐⭐ | ⭐⭐⭐⭐⭐ | 💰💰💰 |
| ☀️GPT-5.4 | ⭐⭐⭐⭐ | ⭐⭐⭐⭐ | ⭐⭐⭐⭐⭐ | ⭐⭐⭐⭐⭐ | 💰💰💰 |
| 🍂DeepSeek V3.2 | ⭐⭐⭐⭐⭐ | ⭐⭐⭐ | ⭐⭐⭐⭐⭐ | ⭐⭐⭐⭐ | 💰 |
| 🎵MiniMax 2.7 | ⭐⭐⭐⭐ | ⭐⭐⭐ | ⭐⭐⭐ | ⭐⭐⭐ | 💰 |
| 🔮Mimo | ⭐⭐⭐ | ⭐⭐ | ⭐⭐⭐⭐ | ⭐⭐⭐ | 💰💰 |
| 💎Gemini 2.5 | ⭐⭐⭐⭐ | ⭐⭐⭐⭐ | ⭐⭐⭐⭐⭐ | ⭐⭐⭐⭐⭐ | 💰💰 |
* Subjective ratings based on Will's real-world usage.
No test history yet