- Alibaba’s Tongyi Lab introduces Qwen Robot Suite
- Its first embedded AI models are divided into navigation (RobotNav), a video “world model” (RobotWorld), and manipulation (RobotManip).
- The move comes after Nvidia recently unveiled and published its own Cosmos 3 offerings.
As much of its AI competition continues to focus on LLMs and making them faster and more capable, Alibaba could be looking to lead on another frontier alongside its LLM ambitions: robots.
The company’s Tongyi Lab has introduced Qwen Robot Suite, what it calls a family of models focused on “embodied AI,” which focuses on enabling machines to perceive space, reason, and act accordingly.
This comes on the heels of Nvidia’s own Cosmos 3, a frontier model for physical AI, further reinforcing CEO Jensen Huang’s narrative that China’s developer ecosystem remains relatively unaffected by chip restrictions, even as the focus in the West continues to shift toward power for many of the sprawling data centers being built in the US.
A competitor or an addition to Nvidia’s playbook?
Qwen-Robot Suite consists of three main models: Qwen-RobotManip, a generalizable vision-language-action model; Qwen-RobotNav, a scalable vision-language navigation model; and Qwen-RobotWorld, a video world model designed for embedded intelligence.
However, there is no denying that robotics is being treated as perhaps the most crucial frontier for AI, even as LLMs continue to advance, with Google and Nvidia among the companies investing billions in research into their respective open source Gemini Robotics and Cosmos offerings.
Alibaba claims that the model, which leverages a lighter Qwen3.5-4B model in place of its Qwen 3.7 Max, which features over a trillion parameters, manages to surpass the real robot benchmark RoboChallenge, with an impressive score of 59.83 and a task success rate of 45%.
With other interested parties such as Tencent, Unitree, AgiBot, UBTech, Galbot, Spirit AI and GigaAI, as well as interest from electric vehicle companies such as Xpeng and Xiaomi, all shaping the future of Chinese AI robotics, R&D in the industry continues in full swing, even as upcoming IPOs are expected to further boost the industry with easier access to capital.
He South China Morning Posta wholly owned subsidiary of Alibaba, noted that “Alibaba’s entry comes at a time when embedded intelligence is rapidly becoming the next frontier of global AI.”
Nvidia’s position on the matter is perhaps more nuanced in attempting to behave as a “facilitator” versus a direct competitor while pushing its open source model to perhaps form the same building block that CUDA makes for GPUs with Cosmos, GR00T, Isaac and similar offerings as a playbook this time to ensure that future robotics platforms are built, like most AI tools, around Nvidia’s hardware and software stack.
Alibaba’s announcement might not be a sign that the Chinese giant is out-engineering Nvidia, but in the context of the Chinese government at least informally insisting on decoupling or, at the very least, not relying on American hardware or software, it can be seen as an intention to build a similar ecosystem for Chinese robotics companies.
In the absence of Nvidia’s presence in China, it could be difficult to compare the two offerings, even though their scales differ considerably: Cosmos 3 is a basic open-world model with multiple scores reported by vendors that don’t cover RoboChallenge, while Alibaba’s are self-reported from exactly one benchmark. Until both approaches can be directly compared, the superiority of one over the other cannot be assumed.
However, what is perhaps painfully obvious to Nvidia, and its CEO has warned time and time again, is that China, irritated by US policies around AI, is no longer looking for open source chips, models or even solutions to incorporate into its ecosystem, but instead wants to build them from scratch.
This could result in a lack of exposure to what was the second most lucrative market for the chip designer, a move that could cost it billions of dollars in revenue in the robotics segment alone, of what is still considered the “world’s factory” due to its huge manufacturing base.
Follow TechRadar on Google News and add us as a preferred source to receive news, reviews and opinions from our experts in your feeds.
