NVIDIA Corporation (NVDA.US) throws 20 billion US dollars to "acquire" its strongest competitor, Wall Street: target price is expected to rise to 300 US dollars.

Wall Street analysts are generally optimistic about the latest deal between NVIDIA Corporation (NVDA.US) and AI inference chip company Groq. Cantor's firm believes that the deal has both "offensive" and "defensive" strategic significance, reiterating NVIDIA Corporation as a "top pick", maintaining a "buy" rating and giving a target price of $300; while Bank of America points out that although NVIDIA Corporation's acquisition of Groq at a high price was unexpected, it successfully transformed the potential ASIC technology threat into its own competitive barrier, making this strategic move significant in the long run, thus maintaining a "buy" rating and a target price of $275. On Wednesday local time, Groq announced that it had signed a non-exclusive licensing agreement with NVIDIA Corporation, allowing the latter to use its inference technology. Under the agreement, Groq co-founder Jonathan Ross, president Sonny Madra, and other team members will join NVIDIA Corporation to drive and expand the deployment of this licensed technology. Reports indicate that NVIDIA Corporation will invest approximately $20 billion to acquire Groq's related assets. Cantor reiterated NVIDIA Corporation as a "top pick", maintaining a "buy" rating and a target price of $300. "On Christmas Eve, NVIDIA Corporation announced the acquisition of Groq's IP and talent for $20 billion, which can be seen as 'talent acquisition'. Overall, we believe that this acquisition has both offensive and defensive attributes. On offense, we understand that NVIDIA Corporation has been working with Groq on specific inference acceleration. We speculate that NVIDIA Corporation sees a real opportunity and believes that making Groq an internal team rather than an external partner is more advantageous," said Cantor analyst team led by C.J. Muse. The analysts also emphasized the strong talent introduction (including the CEO of Groq, who was one of the core developers of Alphabet Inc. Class C (GOOGL.US) TPU). On the defensive side, analysts pointed out that NVIDIA Corporation dominates in AI training and "time-based inference" fields, and Groq's low latency, high energy-efficient inference technology being incorporated into NVIDIA Corporation's complete system stack (likely compatible with CUDA in the future) will help expand its market share in the inference market, especially in the next stage of AI infrastructure development - real-time workloads such as Siasun Robot & Automation, autonomous driving, etc. "Overall, we believe that this acquisition will only strengthen NVIDIA Corporation's full-stack system capabilities and its overall leadership position in the AI market, and further broaden its competitive moat," the Muse team concluded. Bank of America Securities maintains a "buy" rating on NVIDIA Corporation and a target price of $275. Analysts led by Vivek Arya of Bank of America Securities said that the Groq deal was unexpected and costly, but has strategic value. They listed two key points: 1. The Christmas Eve news involving hardware different from the GPUs NVIDIA Corporation excels in - Language Processing Units (LPUs), shows that NVIDIA Corporation realizes that, although GPUs dominate in AI training, the demand for inference is rapidly increasing and may require more specialized chips. 2. Different hardware will increase the complexity of future GPU/LPU roadmaps and pricing, but NVIDIA Corporation can leverage its balance sheet and platform position to provide more choices to customers and conceptually mitigate competition threats from Groq and other specialized ASIC chips. The analysts pointed out that there are still unresolved issues, but in the long term, this potential deal is a positive factor. "In the long run, if this potential Groq deal eventually materializes, its strategic significance may be comparable to NVIDIA Corporation's acquisition of Mellanox in April 2020 - which has now become the foundation of its network/AI expansion moat," the Arya team said. Groq is a star startup specializing in AI inference chips, established in 2016 and headquartered in California, USA. Jonathan Ross, the founder, was a core developer of the Alphabet Inc. Class C's in-house AI chip TPU (Tensor Processing Unit) project. Some former members of the Alphabet Inc. Class C's previous TPU team also joined Groq. Jonathan Ross, as a core developer of the first-generation TPU project at Alphabet Inc. Class C, deeply involved in chip design optimized for AI. In 2016, he led 7 out of 10 core members of the Alphabet Inc. Class C TPU team to leave together and founded Groq. At that time, he found that traditional computing architectures (such as CPU/GPU) were inefficient in handling modern AI tasks, prompting him to start a company that breaks traditional limitations. In February 2024, Groq launched a new AI chip, claiming to achieve "the strongest inference on earth" - running large models on Groq for inference tasks can be 10 times faster or even higher than on NVIDIA Corporation's GPUs. In November 2025, the latest statement from the White House and the U.S. Department of Energy showed that 24 top artificial intelligence companies had signed agreements with the U.S. government to join the "Guangdong Create Century Intelligent Equipment Group Corporation Plan", with both NVIDIA Corporation and Groq included. Groq's core product is the LPU (Language Processing Unit), which is mainly used to accelerate the speed of large language models completing inference-related tasks and is seen as one of the alternatives to NVIDIA Corporation's GPUs. Currently, Groq has partnered with Meta to provide inference acceleration for its Llama API; integrated its AI inference platform with IBM; and signed a huge agreement with Saudi Aramco to build a large AI inference data center.