AI Models
-
Alibaba open-sources Qwen3.6-35B-A3B to rival larger AI models in coding
Alibaba open-sourced the 35-billion-parameter Qwen3.6-35B-A3B model, which activates only 3 billion parameters during each inference.
-
StepFun introduces StepAudio 2.5 TTS, targeting booming AI content market
StepAudio 2.5 TTS is designed to let ordinary users easily become voice directors using natural language.
-
Alibaba launches Happy Oyster world model for real-time 3D creation
Happy Oyster generates dynamic 3D environments, providing technical support for scenarios like film production and game development.
-
Tencent open-sources HY-World 2.0 to turn text into playable 3D worlds
Tencent has officially open-sourced HY-World 2.0, allowing users to generate 3D assets that seamlessly integrate into mainstream game engines using just text or image inputs.
-
Baidu open-sources Ernie-Image model, bringing top-tier rendering to consumer GPUs
Baidu open-sourced its 8-billion-parameter text-to-image model Ernie-Image, which runs smoothly on consumer graphics cards with 24 GB of VRAM.
-
Chinese AI startup MiniMax open-sources self-evolving M2.7 model
MiniMax officially open-sourced its self-evolving M2.7 model, securing deep integration support from leading global chipmakers and inference platforms.
-
Chinese AI star DeepSeek targets late Apr for V4 model release
DeepSeek's next-generation flagship V4 model will be released in late April, according to the Chinese AI company's founder.
-
Zhipu open-sources flagship GLM-5.1 model and raises prices as AI race heats up
GLM-5.1 is currently the only open-source model capable of eight hours of continuous work, Zhipu said.
-
DeepSeek feature update sparks anticipation for V4 model
DeepSeek adds "expert mode," sparking speculation that the underlying model is the upcoming V4 release.
-
Alibaba releases Qwen3.6-Plus, targeting agentic coding
This is one of several closed-source models recently rolled out by Alibaba, as the company seeks to generate more profit from its AI business.
