Alibaba unveils XuanTie C950 RISC-V chip designed for AI agents and cloud workloads

Alibaba delivers RISC-V server chip optimized for Chinese AI

Claims its set performance records but looks to be years behind western fare Alibaba has revealed a new server chip that it says is the most powerful processor ever to use the RISC-V instruction set. According to a social media post by Alibaba's DAMO Academy, which develops some of its chips, the new XuanTie C950 is ready to power cloudy servers, generative AI workloads, high-end robotics, and edge computing devices. "The XuanTie C950 is equipped with a self-developed AI acceleration engine, and for the first time natively supports large models with hundreds of billions of parameters, such as Qwen3 and DeepSeek V3, potentially becoming a new type of high-end CPU for the AI Agent era," the post enthuses. Alibaba claims the machine's single-core general-purpose performance "exceeded 70 points in the SPECint 2006 benchmark test." Photos from Tuesday event at which Alibaba announced the chip suggest its SPECInt 2017 benchmark result is 2.6GHz. Per analysis by Google researcher Laurie Kirk puts it nearly on par with Apple's M1 chip - which the iGiant launched in the year 2020. Kirk also noted that Alibaba claimed it's implemented version 23.1 of the RISC-V RVA, a minor update proposed in August 2025. She expressed surprise Alibaba was able to use it so quickly. A product page and spec sheet for the chip add more details - but also omit details like core count and instead offer the non-specific description of the chip as a "64-bit multi-core CPU IP." Based on those documents the "acceleration engine" mentioned above is probably the XuanTie Tensor Processing Engine (TPE) and supports data types from FP16 down to INT4/FP8, plus the micro-scaling formats MXFP8, MXFP4, and RVFP4. The chip can achieve 8 TOPS per TPE. The memory subsystem apparently "comprises a high-performance multi-level cache hierarchy with an ultralow 4-cycle load-to-use L1 data cache latency, a private per-core L2 cache supporting large capacity configurations, and an MMU with multiple RISC-V virtual memory modes and two-stage address translation." In a section of the spec sheet discussing buses, there's a mention of a multi-processor mode that "leverages the XL-300 interconnect to form clusters of up to 8 cores." Posts The Register has seen suggest Alibaba had the chip made using a 5nm process, a feat that's not beyond some Chinese chipmakers. Whether they can build chips of that caliber in large quantities is less certain. Alibaba CEO Yongming Wu last week acknowledged that Chinese chips lag behind those from western chipmakers. He said the company's response is "to engage in more profound co-design with Alibaba's cloud infrastructure and the Qwen model to provide improved cost effectiveness. This is one key differentiator ... that sets us apart from other chip companies." So while the XuanTie C950 may not be a particularly powerful chip, and Alibaba may struggle to get millions of them into production, native support for the company's own Qwen models suggests Wu's vision for a nicely harmonized AI stack is becoming a reality. ®

Alibaba reveals new AI chip designed for 'agents'

Alibaba on Tuesday announced a new chip designed for agentic capabilities as the Chinese tech giant steps up its semiconductor efforts to fuel its AI push. The XuanTie C950 is a type of chip called a central processing unit (CPU), which Alibaba said will be able to handle the processing of multi-step tasks carried out by AI agents. The term agent refers to an AI system that can carry out a task on behalf of users. The CPU will be installed in data centers and is designed for inferencing, the stage that allows for the actual running of AI models. When it comes to semiconductors and AI, much of the focus so far has been on graphics processing units, or GPUs, a category dominated by tech giant Nvidia. GPUs are critical for training huge AI models because they are able to run multiple calculations at once. Meanwhile, CPUs run general-purpose tasks sequentially, which is key in a world where agents are being designed to carry out specific actions. Alibaba's DAMO Academy, which developed the chip design, said the XuanTie CPUs "can be customized for specific inference patterns, supporting customers in tailoring the chips for their own use." The Chinese tech giant added that when compared to some mainstream products, its CPU achieves over 30% "improvement in performance thanks to its flexibility in customization for specific use cases." The XuanTie C950 is based on RISC-V architecture, which is a rival to the CPU blueprint created by British firm Arm. Companies pay Arm royalties to use its CPU design, whereas RISC-V is effectively a blueprint that can be used for free.

Alibaba develops next-gen chip for agentic AI, Chinese media says - The Economic Times

The 3.2 GHz server chip, built using open-source RISC-V chip architecture, was billed ⁠as "the highest ⁠performing RISC-V CPU in the world" at a conference hosted by DAMO Academy, Alibaba's research arm, according to the reports.Alibaba on Tuesday revealed its next-generation XuanTie C950 5-nanometer processor at an internal conference on Tuesday, local media reported, as the Chinese tech giant gears up for the shift towards agentic AI. The 3.2 GHz server chip, built using open-source RISC-V chip architecture, was billed ⁠as "the highest ⁠performing RISC-V CPU in the world" at a conference hosted by DAMO Academy, Alibaba's research arm, according to the reports. The chip performs more than three times faster than its predecessor, the XuanTie C920, the reports said. The company did not reveal which fab manufactured the chip. Alibaba ⁠did not immediately respond to an emailed request for comment. Alibaba is accelerating in-house chip development through its T-Head ⁠semiconductor arm, primarily focusing on the Zhenwu 810E chip series for AI training and inference, while the XuanTie series is focused on high-performance cloud systems and agentic AI. The move comes after Alibaba last week launched Wukong, its enterprise platform optimised for AI agent workflows, as companies and institutions throughout China adopt OpenClaw. Its international equivalent, Accio Work, was launched on Monday. The agentic AI platform says it can autonomously run complex business operations for small ⁠and medium-sized enterprises. The firm reorganised some of its AI-focused teams under the newly created Alibaba Token Hub earlier this month, which focuses on building AI work platforms for enterprises. The business strategy shift comes as Alibaba finds new ways to ensure profitability as Chinese AI models' token prices have dropped dramatically amid fierce domestic competition.