git
ask
hub
Privacy
Terms
Sign in with GitHub
ghandhitechnology/munbench
↗
Stars ·
1
Language ·
Python
Simplify and Visualize This Repo
Scan the safety of this repo
Self Host this repo
Check out more work of Developer: ghandhitechnology
Ask anything about this repo to start.
Help me set up and run MunBench to evaluate my model's Korean writing ability using OpenRouter, I need to know which API key to get and how to start a full run.
I ran MunBench and got the results. Walk me through interpreting the rubric judge scores, Elo ratings, and judge-free metrics to understand where my model is weak at Korean.
I want to add my own Korean creative writing prompt to MunBench's second track. Show me how to format a new test item and add it to the benchmark pipeline.
Explain how MunBench's three scoring methods work together, rubric judging, Elo comparison, and judge-free tripwires, and which one I should trust most for evaluating Korean prose quality.
Full explanation on explaingit →
📎
Send
By chatting or signing in you agree to the
Terms
and chat-message logging (revocable in
History
).