Xiaomi’s Xring O3 Is a Flex More Than a Threat to Qualcomm, for Now

Xiaomi has unveiled its second homegrown flagship phone chip, the Xring O3, and the framing writes itself: a Chinese giant building its own silicon to loosen its dependence on Qualcomm and MediaTek. That story is true, but it is not quite the whole picture. Look at where the O3 is actually going, and this reads less like an assault on the incumbents and more like Xiaomi proving it can play the game at all.

The chip itself is capable. Built on TSMC’s 3-nanometre process, the O3 packs a 10-core processor and posts benchmark numbers Xiaomi says are the first to clear five million on AnTuTu, with sizeable gains over last year’s Xring O1. It will debut next month inside the Xiaomi 18 Fold and a premium tablet. The company has poured serious money into this, more than 20 billion yuan so far, with plans for 50 billion over the coming decade.

Why the volumes tell the real story

Then you reach the numbers that matter. The O3’s first home, that flagship foldable, has a shipment target of just 200,000 to 300,000 units. For a company that ships tens of millions of phones a quarter, that is a rounding error, and it echoes what happened with the O1: more than a million devices carried it across phones, tablets, and watches, but only around 150,000 were smartphones. Designing a leading-edge chip and buying scarce 3nm capacity to build a few hundred thousand of them is not a cost-saving move. It is a way of demonstrating the capability exists, which is a very different goal from pushing Qualcomm out of the bill of materials.

There is a technical asterisk too. Xiaomi went with 3nm while Qualcomm, MediaTek, and Apple are all moving to 2nm, which leaves the O3 a process generation behind the flagships it is meant to rival. That looks like a deliberate choice to favour yield and reliability over bragging rights, but it is a reminder that catching up is not the same as pulling ahead.

The distinction people keep missing

The most important detail is easy to overlook. Xiaomi is not under the US export restrictions that cut Huawei off from TSMC, so it can still buy the world’s most advanced manufacturing capacity from Taiwan. Its domestic rival has to make do with what Chinese foundries can produce. That means a Chinese company designing an advanced chip in 2026 does not translate into China manufacturing one, a distinction that gets flattened in most coverage but explains how Xiaomi can compete on nodes Huawei cannot reach.

All of this is landing at an awkward moment. Xiaomi’s handset shipments fell to 65 million in the first half of 2026 from 84 million a year earlier, it took the steepest drop among major makers in the June quarter, and rising memory costs are squeezing margins across the industry. Pushing into pricey foldables, where Huawei already holds around 68% of the Chinese market, is an ambitious move in a shrinking one.

So is the Xring O3 a real challenge to Qualcomm and MediaTek? Not this year, and Xiaomi surely knows it. What it is buying is optionality: the in-house expertise, supplier leverage, and product differentiation that only matter years down the line. Every company that eventually made its own silicon a real weapon, Apple included, started with a chip that shipped in tiny numbers and something to prove. This is Xiaomi at that starting line.