Hunyuan
混元大模型
Hunyuan (混元) is Tencent's large model product family, spanning language, image, and video generation. Its best-known recent moment came in October 2025, when Hunyuan Image 3.0 topped the LMArena global leaderboard — beating 26 leading models on real-user preference votes, with 十字路口Crossing noting the model stood out for already carrying "world knowledge" into image generation. On the language-model side, its standing has been more fluid: as of mid-2026, 葬AI's testing still grouped Hunyuan with second-tier Chinese models for instability, though its Hy3 version was described as a striking case of rapid hot-update improvement, climbing within a month from well below DeepSeek's level into the same tier as Qwen 3.7 Max and MiniMax M3. In the wider ecosystem, domestic chipmaker Biren Technology had completed adaptation for the Hunyuan series by April 2026, alongside DeepSeek, Kimi, Qwen, GLM and others.
AI-generated — may contain errors, please verify.
Coverage
Qiming Star | Video Rebirth's Wei Liu: Video Generation Models Are the Optimal Path to Building World Models
Video Rebirth defines the video-native world model as a combination of a world simulator and a world predictor.
Guest Lineup Revealed | Register Now for 2025 WAIC Qiming Venture Partners · Entrepreneurship and Investment Forum — Venture Capital Enters AI Technology and Application Resonance Cycle
This sub-forum will showcase and explore the latest breakthroughs in AI technology and applications through the lens of innovation. Qiming Venture Partners will bring together leading scholars and top entrepreneurs in the field — from both the startup and investment perspectives — for four keynote speeches and four panel discussions.
Looking Back at AI's "May": The Five Most Important Little Things | Attent!on Tech Notes
The future is already here — it's just not evenly distributed.


