VibeCodeArena
An LLM coding arena launched by HackerEarth in which models and human users compete on the same challenges and share a leaderboard, ranked by Elo and evaluation scores.
Timeline (last 90 days)
VibeCodeArena is an LLM coding arena built by HackerEarth, where models and human users tackle the same coding challenges on the same leaderboard.
In the VibeCodeArena leaderboard snapshot, there are 46 models and 100 users in total.
In the VibeCodeArena leaderboard snapshot, DeepSeek-v4-Flash ranks first with an eval score of 95.3 and an Elo of 1639, but its confidence is labeled Low.
In the VibeCodeArena leaderboard snapshot, Claude 3.5 Sonnet ranks 3rd with an eval score of 87.5, Claude Opus 5 ranks 4th with an eval score of 94.0, and Llama-4-Maverick ranks 9th with an eval score of 91.3.
In the Users tab of VibeCodeArena, top users have eval scores around 99.5 and Elo between 1641 and 1642, with some users ranked above the strongest model, DeepSeek-v4-Flash.
VibeCodeArena's Eval score covers dimensions such as security, code quality, correctness, performance and accessibility.
In a specific challenge the author observed, the Metrics tab showed an application with an overall score of 94% that still had 36 major correctness issues.
The Confidence on the VibeCodeArena leaderboard is Low / Medium / High, depending on how much battle data is behind it; Low means few samples and large ranking fluctuation, while High means validated over a large number of matches.
When reading the VibeCodeArena leaderboard, you should look at ranking, confidence and trend together; a High confidence model with an upward trend is more informative than a model sitting at the top with Low confidence.
All Facts (9)
VibeCodeArena is an LLM coding arena built by HackerEarth, where models and human users tackle the same coding challenges on the same leaderboard.
50%UnverifiedIn the VibeCodeArena leaderboard snapshot, there are 46 models and 100 users in total.
50%UnverifiedIn the VibeCodeArena leaderboard snapshot, DeepSeek-v4-Flash ranks first with an eval score of 95.3 and an Elo of 1639, but its confidence is labeled Low.
50%UnverifiedIn the VibeCodeArena leaderboard snapshot, Claude 3.5 Sonnet ranks 3rd with an eval score of 87.5, Claude Opus 5 ranks 4th with an eval score of 94.0, and Llama-4-Maverick ranks 9th with an eval score of 91.3.
50%UnverifiedIn the Users tab of VibeCodeArena, top users have eval scores around 99.5 and Elo between 1641 and 1642, with some users ranked above the strongest model, DeepSeek-v4-Flash.
50%UnverifiedVibeCodeArena's Eval score covers dimensions such as security, code quality, correctness, performance and accessibility.
50%UnverifiedIn a specific challenge the author observed, the Metrics tab showed an application with an overall score of 94% that still had 36 major correctness issues.
50%UnverifiedThe Confidence on the VibeCodeArena leaderboard is Low / Medium / High, depending on how much battle data is behind it; Low means few samples and large ranking fluctuation, while High means validated over a large number of matches.
50%UnverifiedWhen reading the VibeCodeArena leaderboard, you should look at ranking, confidence and trend together; a High confidence model with an upward trend is more informative than a model sitting at the top with Low confidence.
50%