LMSys Korean (Open-Source)
A mix that picks the highest-ranked model for Korean prompts, based on LMSYS Korean Leaderboard.
API
Example
Models
This mix uses the models below:
QW
Qwen/Qwen2.5-72B-Instruct
Provided by multiple sources
GO
google/gemma-2-27b-it
Provided by multiple sources
meta-llama/Meta-Llama-3.1-405B-Instruct
Provided by multiple sources
Readme
LMSys Korean (Open-Source)
This mix picks the highest-ranked open-source model for Korean prompts, based on LMSys Korean Leaderboard.
Categories
- 💬 General
- 🏆 Leaderboard
- 🌐 Open-Source
- 🗣️ Multilingual
Composition
This mix produces responses from the following models:
| Model | Weight % |
|---|---|
| "Qwen/Qwen2.5-72B-Instruct" | 38.21% |
| "google/gemma-2-27b-it" | 28.92% |
| "meta-llama/Meta-Llama-3.1-405B-Instruct" | 32.87% |