Model Kombat by HackerRank
The AI Code Arena
222 followers
The AI Code Arena
222 followers
Coding LLMs go head-to-head on real programming tasks. Developers vote on which solution they'd actually ship. These votes become training data for better models. No synthetic tests. Just code, performance, and brutal honesty.
Incredibly proud of what our team has built! Excited to see which models developers put their trust in the most.
how can I got to know which character refers to which model?
Congratulations on the launch! Are you all using the same questions library from HackerRank?
Model Kombat by HackerRank
@akshita_shah4 we are using them as inspiration. We created a new batch of problem statements for this release!
Congratulations on the launch! I'm curious—are you using the same question library from HackerRank, or are these new ones?
Model Kombat by HackerRank
@akshat_shah14 new ones but using the same rigorous approach we take to creating questions in general
This looks really promising! Congratulations on the launch! 🎉
When are you planning to introduce more models?
Model Kombat by HackerRank
@sanskaragar16 we'll add more models over the next month. For now we wanted to focus on the models that perform best on some popular benchmarks to compare
This is genius, finally a way to benchmark LLMs that actually respects developer standards. “Would you ship this?” is exactly the right question.
Just voted on a Python battle, wild to see how different the solutions were. Love that the voting directly trains the models. This feels like how AI evals should work.