Tesla’s AI6 chip has been finalized using Samsung’s 2nm process, Alibaba and WiMi entering the AI ​​chip fray

According to Korean media reports, details of the collaboration between Tesla (TSLA) and Samsung Electronics (SSNGY) have been further clarified. Tesla’s next-generation core chip, the “AI6,” will officially utilize Samsung’s second-generation 2nm (SF2P) process.

Tesla AI6 chip samples have been revealed.

As an advanced process technology Samsung plans to mass-produce in 2026, SF2P is a key iteration in the 2nm race. Compared to the first-generation 2nm (SF2) process, which is expected to enter mass production in the second half of 2025, it achieves triple optimizations in performance, power consumption, and size.

In addition, the SF2P process offers a 12% performance improvement, better supporting high-computing scenarios; a 25% reduction in power consumption, meeting the low-energy requirements of smart cars and AI devices; and an approximately 8% reduction in chip area, enabling the integration of more transistors within a limited space, further improving chip computing efficiency.

According to the contract, Samsung will exclusively manufacture Tesla’s “AI6” high-performance system semiconductors. This chip will be directly used in Tesla’s three core next-generation businesses: Fully Self-Driving (FSD) systems, humanoid robots, and data centers.

Alibaba is developing a new AI chip.

According to media reports, Alibaba (BABA) is developing a new AI chip aimed at filling the gap left by Nvidia in the Chinese market. Currently in the testing phase, the chip is primarily targeted at a wider range of AI inference tasks and is compatible with Nvidia’s CUDA architecture.

Guosheng Securities noted that Alibaba has established full-stack AI technical capabilities, which is expected to correct the gap in cloud service expectations and initially close the loop in domestic computing power.

Kaiyuan Securities believes that over the next three years, Alibaba will invest over 380 billion yuan in building cloud and AI hardware infrastructure. AI is driving Alibaba Cloud into a positive cycle, while Alibaba’s high capital expenditure growth is expected to drive a fundamental turning point in the domestic AI computing power chain.

Demand for AI continues to surge.

Notably, Nvidia (NVDA)’s new quarterly financial report showed both revenue and profit exceeding market expectations, and stated that sales growth this quarter will remain above 50%. In its latest earnings call, NVIDIA highlighted the progress of core products such as Blackwell and Rubin. The Blackwell platform achieved record performance, with a 17% quarter-over-quarter increase. NVIDIA officially launched mass production of the GB300 chip in the second quarter.

The Rubin chip, on the other hand, has entered the wafer fabrication phase and remains on track for mass production next year. It will become NVIDIA’s third-generation NVLink rack-scale AI supercomputer, and the supporting supply chain is mature. Shipments of the H100 and H200 chips, based on the Hopper architecture, increased in the second fiscal quarter.

Industry insiders point out that NVIDIA is fully committed to responding to surging global demand, and its AI infrastructure solutions will play a key role. With the continued acceleration of global AI computing infrastructure deployment and continued strong demand, coupled with the expected continued shipments of NVIDIA’s Blackwell architecture GPUs and the Rubin architecture cabinets, the entire AI chip market still has significant room for growth.

WiMi Joins the AI ​​Chip Race

Information indicates that Wimi Hologram Cloud Inc. (WIMI), a rapidly growing tech startup, has shown a strong optimism about the computing power sector’s industry trends. In recent years, it has actively participated in the global AI chip race by building a computing power base, participating in an open source ecosystem, and developing cutting-edge technologies. Its strategy encompasses hardware adaptation, ecosystem collaboration, and application deployment, meeting diverse computing power needs.

In fact, by establishing an open source AI chip ecosystem, WiMi is reducing access costs for small and medium-sized enterprises and accelerating the commercialization of scientific research results. For example, the open source technology commercialization model it promotes optimizes computing power adaptation for scenarios in industries such as industry, healthcare, and finance, improving the performance of AI models in low-computing environments.

In addition, WiMi has focused on optimizing cloud computing parallelism. Regarding its inference platform, WiMi continues to deepen its focus on core business scenarios. Following early breakthroughs in large-scale cloud computing optimization and edge computing performance optimization, WiMi has further innovated, significantly reducing quantum communication latency and improving communication and computing parallelism efficiency, achieving industry-leading performance across all metrics.

Summary
Driven by the explosive growth of AI applications, the chip industry is experiencing development opportunities. As AI and intelligent agents gradually permeate various industries, the surge in AI demand is boosting the chip industry, including high-end computing chips, edge AI chips, and analog chips. This series of changes reflects not only a restructuring of the supply chain but also a comprehensive game of technological advancements, market structure, and even national power. Therefore, the AI ​​industry is expected to show a positive trend of rising demand and optimistic expectations, and opportunities in the related computing power sector are attracting significant attention.