Cambricon doubles quarterly revenue as demand for AI chips booms in China
Cambricon has sharply strengthened its position in China's AI chip market: in the first quarter of 2026, revenue rose to 2.88 billion yuan and net profit reached 1.01 billion yuan. The growth is fueled by Beijing's push for technological self-reliance, strong demand for computing power, and large customers shifting toward local accelerators. After the earnings were published, the company's shares in Shanghai jumped 14%.
AI-processed from Bloomberg Tech; edited by Hamidun News
Chinese AI chip developer Cambricon sharply accelerated its growth in the first quarter of 2026: revenue more than doubled, and profit grew even faster. For the Chinese market, this is an important signal: the policy of technological self-reliance is turning into more than slogans, and is now producing real sales of local accelerators.
A quarter of breakout growth
Cambricon reported revenue of 2.88 billion yuan for January–March 2026, against 1.11 billion yuan a year earlier. That is growth of roughly 160% year over year. Net profit for the quarter was 1.01 billion yuan, versus 356 million a year ago, an increase of about 185%.
The market reacted quickly: after the earnings release, the company's shares in Shanghai jumped 14%, their strongest daily gain since late August.
The quarter also looks like part of a longer trend rather than a one-off spike. In February, the company reported its first annual profit after several loss-making years, and in March it announced a dividend of 15 yuan per 10 shares. The new report shows that Cambricon did not simply turn profitable on the back of the AI frenzy, but managed to lock in that turnaround at the start of 2026.
Operating cash flow for the quarter reached 834 million yuan, and total assets at the end of March were up 14.6%. The report also shows a financial cushion: gross margin was 54.33%, and substantial R&D spending continued even amid explosive sales growth. That matters because the AI chip market is won not only on current shipments, but also on the speed of the next product cycle.
Why demand is growing
The key reason is simple: China needs its own computing capacity to train and run AI models, and dependence on foreign suppliers is becoming increasingly risky. Beijing continues to pursue self-sufficiency in critical technologies, including semiconductors, and restrictions and uncertainty around US chip supplies only increase demand for local alternatives. Against this backdrop, Cambricon is winning orders where foreign solutions dominated until recently.
Cambricon attributed the jump to "sustained growth in demand for
computing power in the AI industry."
Here is how the quarter's numbers look:
- revenue: 2.88 billion yuan, up 159.6%
- net profit: 1.01 billion yuan, up 185%
- profit excluding one-off items: 934 million yuan, up 238.6%
- operating cash flow: 834 million yuan, up 159.6%
- R&D spending: 324 million yuan, or 11.23% of revenue
The report matters also because it shows a reshuffling of China's AI hardware landscape as a whole, not just the success of one company. Another local developer, MetaX, also reported strong quarterly revenue growth, up 75% year over year. But Cambricon now looks like one of the main beneficiaries of the shift to domestic suppliers: it already has scale, access to public capital markets, and the status of a prominent candidate to be China's replacement for Nvidia in AI accelerators. Competition has not gone away, though: other players are growing quickly in the country, including projects built around local GPUs and specialized server chips.
What this means
Cambricon's results show that the substitution of imported AI chips in China has moved from strategy to revenue, profit and stock market reaction. If demand for local accelerators holds, the company could become one of the key infrastructure suppliers for China's AI market.
The main question now is not whether demand exists, but how quickly the domestic ecosystem can scale production, memory, packaging and software around such chips.
- Nov 1, 2026Local models lose to Opus and Codex: 0.21 on a MacBook versus 1.00 for Codex
- Nov 1, 2026Habr author fine-tunes YOLOv8l on dental X-rays and pushes mAP50 to 0.642
- Nov 1, 2026Voice bots built on LLMs turn to Jev, a fast classifier, for decisions in 200–500 ms
- Nov 1, 2026MoneyGate safety layer blocked all 42 fraudulent requests to local AI models
Key Questions Answered
What are the key details on A quarter of breakout growth?
Cambricon reported revenue of 2.88 billion yuan for January–March 2026, against 1.11 billion yuan a year earlier. That is growth of roughly 160% year over year. Net profit for the quarter was 1.01 billion yuan, versus 356 million a year ago, an increase of about 185%. The market reacted quickly: after the earnings release, the company's shares in Shanghai jumped 14%, their strongest daily gain since late August. The quarter also looks like part of a longer trend rather than a one-off spike. In February, the company reported its first annual profit after several loss-making
Why demand is growing?
The key reason is simple: China needs its own computing capacity to train and run AI models, and dependence on foreign suppliers is becoming increasingly risky. Beijing continues to pursue self-sufficiency in critical technologies, including semiconductors, and restrictions and uncertainty around US chip supplies only increase demand for local alternatives. Against this backdrop, Cambricon is winning orders where foreign solutions dominated until recently. Cambricon attributed the jump to "sustained growth in demand for computing power in the AI industry." Here is how the q
What this means?
Cambricon's results show that the substitution of imported AI chips in China has moved from strategy to revenue, profit and stock market reaction. If demand for local accelerators holds, the company could become one of the key infrastructure suppliers for China's AI market. The main question now is not whether demand exists, but how quickly the domestic ecosystem can scale production, memory, packaging and software around such chips.
Test Frontier LLMs & Autonomous Agents in One Workspace
Run Claude Opus 5.5, GPT-6, Gemini 3.1, and DeepSeek side-by-side in Nomad AI (nomadnet.ai) — unified model routing, deep research, and multimodal workflows under one subscription.
The AI world, distilled — once a week
Seven stories that actually mattered, hand-picked. No noise, no reposts, no press releases.
Done! Check your inbox for a confirmation.