Key Info

Tencent Hunyuan's Hy4 is now ranked the #1 Chinese model on Agent Arena, an evaluation built on more than 12,000 real-world, long-horizon agentic sessions from global users. It also places Tencent Hunyuan as the only Chinese lab in the top 5, with a +5.4% net improvement.

Highlights

  • Hy4 holds the top spot among Chinese models on Agent Arena.
  • The ranking draws on over 12K real-world, long-horizon agentic sessions rather than a synthetic test set.
  • Tencent Hunyuan is the only Chinese lab in the top 5 of the leaderboard.
  • Hy4 shows a +5.4% net improvement in the evaluation.
  • The team frames the model as built for task queues and real work rather than benchmark tuning.