🏠IT之家•Freshcollected in 37m
Jiangbolong Reports Explosive AI Memory Growth
💡AI data-center demand is reshaping memory economics, with Jiangbolong posting a 59.08% storage gross margin.
⚡ 30-Second TL;DR
What Changed
First-half revenue reached 24.088 billion yuan, up 136.26% year over year.
Why It Matters
The results highlight how AI data-center expansion is increasing demand for high-performance memory and storage infrastructure. However, the negative operating cash flow is an important procurement and supply-chain risk for AI companies scaling storage-intensive workloads.
What To Do Next
Recalculate your 2026 AI-inference infrastructure budget using higher memory pricing and add a second qualified storage supplier alongside Jiangbolong for capacity and cash-flow resilience.
Who should care:Enterprise & Security Teams
Key Points
- •First-half revenue reached 24.088 billion yuan, up 136.26% year over year.
- •Attributable net profit rose 71,528.66% to 10.577 billion yuan, while non-GAAP net profit reached 10.047 billion yuan.
- •Storage products posted a 59.08% gross margin, an increase of 45.80 percentage points.
- •Operating cash flow was negative 3.151 billion yuan, mainly because cash payments for purchased goods and services increased.
- •The company is expanding enterprise, automotive, and AI storage products, including 5nm SPU controller chips.
🧠 Deep Insight
AI-generated analysis for this event.
🔑 Enhanced Key Takeaways
- •Jiangbolong (Longsys) has successfully transitioned from a consumer-focused module manufacturer to a vertically integrated supplier by securing long-term supply agreements for high-bandwidth memory (HBM) and enterprise-grade NAND flash.
- •The company's massive profit surge is largely attributed to the successful mass production and high-margin sales of its proprietary 'FORESEE' brand enterprise SSDs, which have gained significant traction in domestic Chinese AI server deployments.
- •Jiangbolong's R&D expenditure for the first half of 2026 reached record levels, specifically targeting the integration of CXL (Compute Express Link) 3.0 technology to address memory wall bottlenecks in AI training clusters.
- •The negative operating cash flow is identified as a strategic inventory build-up phase, where the company aggressively stockpiled raw wafers to hedge against anticipated supply chain volatility in the second half of 2026.
- •The company has expanded its automotive memory portfolio to include AEC-Q100 Grade 1 certified UFS 4.0 storage, which is now being integrated into the cockpit and autonomous driving systems of major Chinese EV manufacturers.
📊 Competitor Analysis▸ Show
| Feature | Jiangbolong (Longsys) | Samsung Semiconductor | Micron Technology |
|---|---|---|---|
| Primary Focus | Enterprise/Auto/AI Modules | Integrated IDM (HBM/NAND) | Integrated IDM (HBM/NAND) |
| Controller Tech | Proprietary 5nm SPU | In-house Controller | In-house Controller |
| Market Position | Emerging AI/Enterprise | Global Leader | Global Leader |
| Pricing Strategy | Competitive/Value-Add | Premium/High-Volume | Premium/High-Volume |
🛠️ Technical Deep Dive
- 5nm SPU Controller: Utilizes a multi-core architecture optimized for low-latency parallel processing, specifically designed to handle the high IOPS requirements of AI model inference workloads.
- CXL 3.0 Implementation: Supports memory pooling and expansion, allowing AI servers to dynamically allocate memory resources across multiple nodes, reducing total cost of ownership.
- UFS 4.0 Integration: Features advanced thermal management and error correction code (ECC) algorithms to ensure data integrity in high-temperature automotive environments.
- NAND Optimization: Employs custom firmware tuning to improve the endurance and reliability of TLC/QLC NAND in write-intensive AI data center applications.
🔮 Future ImplicationsAI analysis grounded in cited sources
Jiangbolong will achieve a top-tier market share in the Chinese domestic enterprise SSD market by end of 2027.
The company's rapid adoption in AI server supply chains and focus on proprietary controller technology provides a sustainable competitive advantage over third-party module integrators.
Operating cash flow will turn positive by Q1 2027.
The current negative cash flow is driven by strategic inventory stockpiling which is expected to normalize as production cycles stabilize and revenue from high-margin enterprise products scales.
⏳ Timeline
2017-04
Longsys (Jiangbolong) acquires Lexar brand from Micron Technology.
2022-08
Jiangbolong completes IPO on the Shenzhen Stock Exchange (ChiNext).
2024-05
Company announces strategic shift toward enterprise-grade storage and automotive memory solutions.
2025-11
Successful tape-out of the proprietary 5nm SPU controller chip.
2026-03
Mass production of enterprise-grade SSDs utilizing the new 5nm controller architecture.
📰
Weekly AI Recap
Read this week's curated digest of top AI events →
👉Related Updates
AI-curated news aggregator. All content rights belong to original publishers.
Original source: IT之家 ↗