Show HN: ModelKombat – Arena-style battles for coding models(astra.hackerrank.com)
astra.hackerrank.com
Show HN: ModelKombat – Arena-style battles for coding models
https://astra.hackerrank.com/model-kombat
4 comments
When I tried this, one of the two C++ models kept generating pages and pages of ideas for how it could implement the code, but just got stuck in a loop doing this and failed to actually implement anything.
Then again, I’ve seen interviews play out like that too.
Then again, I’ve seen interviews play out like that too.
Super interesting to see an arena specific to coding. I love the retro throwback.
How are you deciding which challenges to start with for this first release?
How are you deciding which challenges to start with for this first release?
Over the years, we have built up deep expertise in generating programming challenges, and we are now using that to make coding models better.
Our first launch is Model Kombat -- an arena where you can directly compare anonymized coding models, side by side, on real problems.
* Pick an arena (Java, Python, etc.) * Each battle has 3 rounds: see the problem + two model outputs -> vote on which you’d actually prefer * Leaderboards + problem statements are updated weekly.
We already have 400+ challenges live, and our goal is to evolve toward real-world, multi-file challenges and robust tooling for coding model evaluation.
Try it now at: modelkombat.com
We’d love your feedback.