Longsys Q1 FY26 — The Memory Upcycle Lands on a Module Maker
Revenue more than doubled and the bottom line swung from a loss to a RMB 3.9B profit in one quarter. The AI memory shortage has reached the layer that buys wafers rather than makes them — which is exactly why the next two prints matter more than this one.
Published · Post-print analysis · 301308.SZ · For analysts covering memory, storage, and AI infrastructure
The setup
Longsys isn't a fab. It's a memory module and storage maker — it buys DRAM and NAND wafers from the upstream giants, packages and controls them, and sells finished product under Lexar (retail/channel), Zhitai (致态, China consumer SSD) and an enterprise/embedded line. That model is the whole story of this print. When wafer prices rip, a packager's input costs rise — but in a genuine shortage, its finished-goods ASPs and the carrying value of pre-secured inventory rip faster. Q1 FY26 (calendar Jan–Mar 2026, reported alongside the FY2025 annual report) is what that looks like when it works in the company's favour.
The backdrop is the cleanest memory upcycle since 2017–18: AI datacenter buildout is vacuuming up HBM and high-density DRAM/NAND, the leading suppliers have tilted capacity toward AI parts, and commodity/legacy supply has tightened as a second-order effect. Into that tape, a downstream packager just posted numbers that look like upstream numbers.
Where Longsys's Q1 2026 revenue yuan went: 55.5% survives cost of sales, 47.5% survives the other operating costs, 38.1% lands as profit before other income — on a gross margin that went to 55.5% from 10.4% and a quarter that turned a year-ago loss into RMB 3,862M for the owners of the parent. RMB 9,909M − RMB 4,407M = RMB 5,502M gross profit; less RMB 793M of the remaining costs the filing groups under total operating cost = RMB 4,709M. That is not the filing's operating-profit subtotal of RMB 4,762M, which is struck after impairments, other income, investment income and fair-value changes; the chart puts each of those below the operating line instead. Inflows there — RMB 199M other income, RMB 6M gains on asset disposal, RMB 0M non-operating income — cannot be drawn against a flow, so the chart shows profit excluding them: RMB 4,709M − RMB 935M = RMB 3,774M; that plus RMB 205M = RMB 3,979M net profit, of which RMB 3,862M is attributable to owners of the parent and RMB 117M to minority interests. Finance costs of RMB 155M sit inside operating cost, as the filing groups them. Y/Y on revenue against RMB 4,256M in Q1 2025. Three months ended March 31, 2026.
The numbers
Operating revenue of RMB 9.909 billion (≈US$1.38B), up 132.8% year-on-year and 64.3% sequentially. Net profit attributable to shareholders of RMB 3.862 billion (≈US$0.54B), a turnaround from a RMB 152 million loss in Q1 FY25 and up 443% QoQ. Stripping non-recurring items, net profit was RMB 3.943 billion versus a RMB 202 million loss a year earlier — the ex-items figure being larger than the headline tells you the operating business, not one-offs, did the work. That's roughly RMB 9.2 of EPS in a single quarter against a ~419M share count.
For run-rate context: trailing-twelve-month revenue is now ~RMB 28.4B with ~RMB 5.4B of net income (~32% gross margin, ~19% net margin TTM), against FY2025's full-year RMB 22.77B revenue / RMB 1.42B net profit. A single quarter just contributed more than a quarter of TTM revenue and the majority of TTM profit — the signature of a cyclical inflecting hard, not a steady compounder.
Paid subscribers
The rest of this dive is for paid subscribers.
The headline numbers are above. The 6 sections below carry the
mechanism, the peer read, the valuation work, and the dated tests that decide it.
Every Monday, the single sharpest call from the pipeline — a dated, falsifiable read on credit, demand, and capex, then graded in public on the scorecard. Free.
USD conversions approximate (~7.18 CNY/USD); segment-level (Lexar / Zhitai / enterprise) revenue split not broken out in the summary release — unverified pending the full report.
