AITOPAI热报
ModelsMulti-source

腾讯 Hunyuan Hy4 登顶 Agent Arena 中国模型榜首

Thanks @arena! 🎉 Really cool to see Hy4 as the top Chinese model on Agent Arena.

Hunyuan@TXhunyuan
Post· 9 sources
View the post
Why we picked it

腾讯的 Hy4 在 Agent Arena 智能体榜单拿了中国模型第一，12000 多场真实任务测出来的，单任务成本才 0.57 美元。

腾讯 Hunyuan 的 Hy4 在 Agent Arena 榜单上成为排名第一的中国模型，也是唯一进入前 5 的中国实验室。该排名基于超过 12000 场真实用户的长程智能体任务会话。Hy4 相比此前有 5.4% 的净提升，每个任务的中位成本为 0.57 美元，处于榜单的 Pareto 前沿。

Original · Hunyuan

Thanks @arena! 🎉 Really cool to see Hy4 as the top Chinese model on Agent Arena.

Thanks @arena ! 🎉 Really cool to see Hy4 as the top Chinese model on Agent Arena. Arena.ai @arena Hy4 by @TencentHunyuan is now officially the #1 best Chinese model in Agent Arena! With over 12K+ real-world, long-horizon agentic sessions from our global community of users, Hy4 has also placed Tencent Hunyuan as the only Chinese lab in the top 5. It is now showing +5.4% net improvement, and sits firmly on the Agent Arena Pareto frontier at a $0.57 median cost per task. Congrats again to the @TencentHunyuan team! 🔗 View Quoted Tweet 💬 4 🔄 2 ❤️ 42 👀 2321 📊 8 ⚡

  • lmarena.ai11-04 07:15原文
  • 掘金本周最热11-04 08:14原文
  • Artificial Analysis11-05 08:28原文
  • kimmonismus11-04 06:53原文
  • shao__meng11-04 13:05原文
  • Nous Research11-05 00:21原文
  • CNA: Business11-04 07:26原文
  • Nikkei Asia11-04 09:27原文
#Tencent Hunyuan#Hy4#Agent Arena#智能体#推理模型

Keep reading

Recent stories with the same tag or section
ModelsBuildersOfficial71°10:20
Chinese content腾讯 Hunyuan 发布 Hyra 2.0 科研智能体，可自建工具用于蛋白质设计、数学证明与量子化学

腾讯 Hunyuan 的 Hyra 2.0 不只会答题，还能自己造工具：蛋白质设计快 57 倍，连数学证明和量子化学代码都搞出来了，开源可看。

#Hyra 2.0#Tencent Hunyuan#智能体#蛋白质设计
Event
官方一手Pandaily@[email protected] (Pandaily)11 sourcesSource
ModelsBuildersSelected01:02
Tencent Hunyuan releases Hy-Nomas open source interactive video world model

Consumer-GPU inference lowers the hardware barrier that keeps most world models out of hobbyist and lab budgets, and LoRA hooks make it a practical base for robotics simulation. Developers building interactive video, embodied AI or game prototyping can grab the weights from Hugging Face and fine-tune on their own data. Compared with closed world models behind APIs, Hy-Nomas gives full control over generation pipelines.

#Hunyuan#Hy-Nomas#Tencent#video generation
官方账号腾讯混元@TencentHunyuanSource
PapersBuildersSelected17:16
Tencent Hunyuan releases WorkForge for training work agents

The dual programmatic-plus-semantic verifier design addresses the hardest part of agent training: rewarding long-horizon tasks without human grading. Teams building enterprise work agents can reuse the recipe to generate their own verifiable task suites from internal workflows rather than hand-crafted benchmarks. It sits alongside agent-environment efforts like task synthesis pipelines from other labs but anchors everything in expert-derived, evidence-backed tasks.

#Tencent Hunyuan#WorkForge#agents#open-source models
官方账号腾讯混元@TencentHunyuanSource
ModelsBuildersSelected17:09
Tencent Hunyuan releases WorkForge agent training framework

Training agents on verifiable, evidence-grounded tasks addresses the core bottleneck of RL for long-horizon work: reward reliability. The 16.7K-environment scale makes this one of the larger open efforts in agent environment construction. Engineers building work or office agents can study the dual-verifier design for their own training pipelines.

#Tencent Hunyuan#WorkForge#agents#AI agents
官方账号腾讯混元@TencentHunyuanSource