Claude Fable 5 has lost its top position in the AI coding leaderboard. Kimi-K3, built by China’s Moonshot AI, now holds the first place with 1679 points, while Fable 5 scored 1679 points, according to Arena’s latest rankings.
How Kimi-K3 Beat Claude Fable 5
The win came unexpectedly, with Moonshot’s previous model, Kimi-K2.6, ranking 18th. Kimi-K3 shot straight to first, surpassing Claude Fable 5 by 48 points. Kimi-K3 also won six of seven coding categories, covering marketing pages, data dashboards, and consumer apps. Claude Fable 5 kept first place in just one category, Gaming.
Why This Matters Beyond One Leaderboard
Kimi-K3 undercuts its rival on price, with Moonshot listing it at $3 per million input tokens and $15 for output, compared to Arena’s listing of Fable 5 at $10 and $50. Moonshot will also make Kimi-K3’s full model weights available by July 27, allowing anyone to run the top coding model for free. The rivalry between AI models is heating up, with Chinese models overtaking US rivals in monthly token use. Alibaba has also told staff to drop Claude Code over security fears. The next few weeks will show whether Kimi-K3 keeps its lead as more votes come in.
Based on reporting from crypto.news.
