# Huawei Ascend
Latest news and articles about Huawei Ascend
Total: 19 articles found
The Silicon Siege: Why China’s AI Champions are Racing to Build Their Own Chips
Zhipu AI is reportedly moving into chip design to mitigate high inference costs and US-led supply chain restrictions. This shift toward vertical integration reflects a broader trend among Chinese AI leaders seeking to secure compute sovereignty through custom, optimized silicon.
The Trillion-Parameter Breakthrough: Meituan’s Open-Source Milestone Signals Maturity for China’s AI Hardware
Meituan has open-sourced LongCat-2.0, a trillion-parameter AI model that is the first of its scale to be trained and inferred entirely on Chinese domestic hardware. This release, synchronized with major chipmakers like Huawei and Moore Threads, signals a significant step toward China's goal of technological self-sufficiency and digital sovereignty.
China’s Silicon Sovereignty: Meituan’s LongCat 2.0 Signals a Turning Point for Domestic AI Clusters
Meituan has launched LongCat 2.0, a 1.6-trillion parameter AI model trained entirely on a domestic 50,000-card computing cluster. The achievement signals a major shift in China's AI industry toward hardware independence and strategic integration of AI into core business operations.
DeepSeek’s Vision Quest: Multimodal Ambitions Meet the ‘Founder Recognition’ Glitch
DeepSeek has launched its first multimodal image recognition feature, marking a significant upgrade to its AI ecosystem. While the tool currently suffers from humorous identification glitches involving its founder, a recent $7 billion funding round and tight integration with Huawei’s domestic chips solidify its position as a top-tier contender in China’s AI landscape.
Breaking the Silicon Ceiling: Zhipu AI’s GLM-5.2 and the Dawn of China’s Sovereign Compute Ecosystem
Zhipu AI has launched and open-sourced GLM-5.2, a high-performance model that leads global coding benchmarks. Crucially, the model is fully optimized for a wide range of domestic Chinese chipsets, signaling a major step toward an independent AI technology stack.
DeepSeek’s $50 Billion Reality Check: China’s AI Maverick Swaps Idealism for Capital
DeepSeek founder Liang Wenfeng is abandoning his 'no-funding' stance to raise $7.4 billion at a $50 billion-plus valuation to combat talent poaching and high compute costs. The move marks a strategic shift toward commercialization and a deep integration with Huawei’s domestic hardware ecosystem as the Chinese AI market enters a phase of total competition.
The Rise of the 'Token Factory': Huawei and Hongxin Electronics Pioneer a Domestic AI Computing Hub in Wuxi
Hongxin Electronics and Wuxi High-tech Zone have partnered to launch a 'Token Factory' powered by Huawei Ascend 384 super-nodes. This infrastructure project aims to provide high-density domestic computing power to support the growing demands of China's generative AI sector.
From Spirits to Silicon: Cambricon Reclaims China’s Stock Market Throne Amid AI Gold Rush
Cambricon Technologies has reclaimed the title of China's most expensive stock following an explosive Q1 2026 earnings report characterized by 159.6% revenue growth and positive cash flow. The surge reflects the massive demand for domestic AI hardware, though the company faces risks from high customer concentration and intensifying price competition with Huawei and Nvidia.
The DeepSeek Disruptor: China’s AI Upstart Wages a Brutal Price War on Silicon Valley
DeepSeek V4 has launched a major challenge to Western AI dominance by offering high-tier performance at a radical discount, often costing 30 times less than GPT-5.5. While it struggles with the most complex autonomous engineering tasks, its efficiency and deep integration with domestic Chinese hardware like Huawei’s Ascend chips signal a shift toward a self-reliant and hyper-competitive Chinese AI ecosystem.
DeepSeek Defies the Odds: A Cutthroat Price War Ignites in China’s AI Market
DeepSeek has triggered a massive AI price war in China by slashing token prices by 75%, even as giants like Baidu and Tencent raise fees due to hardware costs. By leveraging a new high-efficiency architecture and optimizing for domestic hardware like Huawei's Ascend chips, DeepSeek aims to disrupt the market economics of generative AI.
The DeepSeek Paradox: China’s Latest AI Model Crowns Open-Source Rankings Despite Staggering Hallucination Rates
DeepSeek-V4 has emerged as a top-ranked open-source AI agent, yet overseas testing reveals a critical 96% hallucination rate. While the model bolsters China's domestic semiconductor stocks through hardware compatibility, its factual reliability remains a significant hurdle for global adoption.
DeepSeek’s V4 Gambit: Low-Cost LLMs and the Pivot to Domestic Powerhouses
DeepSeek has launched its V4 model series, featuring a 1.6 trillion parameter flagship that optimizes million-token contexts at unprecedented price points. The release highlights a deepening partnership with Huawei's Ascend compute platform, signaling a strategic shift toward domestic hardware autonomy amidst global supply constraints.