Huatian is the third of the three mainland OSATs and the one that grew fastest in the first half of 2026. Revenue was up 35%, ahead of both larger competitors. That makes it worth a close look, and the close look has a caveat attached.
The caveat is not a criticism. It is the difference between a recovery and a structural change, and it matters for how you read the line's capacity.
The half-year numbers
H1 2026 revenue was 10.51 billion RMB, up 35.09% year-on-year. Second-quarter revenue was 5.71 billion RMB, up 912 million sequentially and described as a single-quarter record. Attributable net profit was 813 million RMB, up 259.15%, of which 727 million came in the second quarter.
Ex-items attributable net profit was 250 million RMB. That figure is the one to anchor on: the reported 813 million includes a large non-operating component, and the ex-items number is what the assembly business itself produced.
Both directions are real. A swing from a thin or negative ex-items position to 250 million RMB is a genuine recovery driven by utilisation. It is not yet evidence of a structurally higher margin business.
Five bases, split by package class
Huatian runs a distributed multi-base network with a clear division of labour:
- Tianshui: legacy package families — DIP, SOT, SOP, QFP
- Xi'an: memory, image sensors, power devices
- Kunshan: FC-BGA, fan-out, wafer-level packaging and SiP
- Nanjing: server memory packaging, including DDR5 capacity reported as ramping
- Malaysia: overseas orders, providing geographic hedging
Why that structure matters for cost
The split matters because it is more explicit than the other two. Legacy and advanced capacity are separated by site rather than competing for the same line, which is part of why Huatian can quote competitively on established packages while building advanced capability elsewhere.
Where the growth came from
Secondary coverage attributes the first-half performance largely to storage: server memory packaging capacity at Nanjing ramping as memory pricing recovered, with volume and price both reported as improving and utilisation and margin repairing as a result. Advanced packaging routes including 2.5D and FC-BGA are reported as having entered production, and automotive power device packaging as benefiting from electric vehicle volume.
Those attributions come from financial media rather than from the company filing, and they are recorded here as such. The company's own product list spans DIP, SOT, SOP, QFP, QFN/DFN, BGA/LGA, FC, MCM, SiP, WLP, TSV, bumping, MEMS, fan-out, PLP and 2.5D/3D.
The cyclicality you are buying into
Huatian's growth is more levered to the storage cycle than either of the other two. That is why it grew fastest in an upswing, and it is also why the same business will feel a downturn more sharply. Advanced packaging capacity carries heavy depreciation, which pressures margin while a line is ramping.
For a programme this cuts both ways. In an upcycle, Huatian has both capacity and motivation to take new work. In a tight market, the same exposure means its queue position can change faster than a more diversified competitor's.
Which programmes fit Huatian
Cost-driven programmes with an established package form factor are the natural fit — the legacy base is real capacity with real cost discipline behind it. Automotive and sensor content, and storage-adjacent packaging, are the next best fits given where capacity sits.
It is a weaker fit for exploratory advanced packaging where the route is not yet settled. That is the case for a broader platform. And if your programme is schedule-critical into a storage upswing, ask specifically how your volume band is prioritised at Kunshan and Nanjing.
