This mix picks the highest-ranked model for Korean prompts, based on LMSys Korean Leaderboard.
- 💬 General
- 🏆 Leaderboard
- 🗣️ Multilingual
This mix produces responses from the following models:
| Model | Weight % |
|---|---|
| "gemini-1.5-pro-002" | 72.01% |
| "chatgpt-4o-latest" | 16.92% |
| "gemini-1.5-pro-exp-0827" | 7.64% |
| "gemini-1.5-pro-exp-0801" | 3.42% |