China Hot Tech @ChinaHotTech
← All posts
Chips · AI Infra · 3D Packaging

Dongfang Suanxin DF1000: 3D Near-Memory Chip Cluster Without HBM

Dongfang Suanxin (东方算芯 / Oriental Compute Chip), a Shanghai Zhangjiang startup founded May 2024 and chaired by Tsinghua-pedigree semiconductor figure Wei Shaojun (魏少军), is pitching the DF1000 as a software-defined near-memory 3D AI chip: a DRAM wafer bonded atop a logic wafer via Cu-Cu hybrid bonding — no HBM stacks, no bumps/interposer* in the vendor story — on a 14nm process. Chinese press covering Shanghai’s 国芯展 reports that the first domestically fully-sourced large-scale training cluster built on DF1000 has been delivered: 128 cards, continuous training for 72 hours without interrupt*. Chip launch coverage dates to July 13, 2026; cluster-delivery pieces updated around Oct 30–31. Every FLOPS, bandwidth, and uptime figure below is a vendor or press claim — flagged with asterisks.

· ~7 min read · China Hot Tech

Soft sand / copper / slate editorial: Dongfang Suanxin DF1000 bonded DRAM-on-logic wafers with 6.4 TB/s bandwidth arrows and HS128 cluster callouts

Illustration: soft sand + copper/bronze + cool slate — two wafers bonded, bandwidth arrows, cluster chips. Asterisked labels are vendor/press figures, not third-party audits.

Training clusters have a bandwidth problem as much as a FLOPS problem. HBM is scarce, expensive, and geopolitically concentrated; process nodes below 7nm are constrained for Chinese fabs. DF1000’s bet is a third path between custom ASICs and GPGPU copies: keep logic at a reachable 14nm node, then recover memory bandwidth by stacking DRAM directly on the compute die with wafer-level hybrid bonding. Vendor numbers put peak BF16 compute at 520 TFLOPS*, on-chip memory bandwidth at 6.4 TB/s*, and scale-up interconnect at 900 GB/s*. Whether those envelopes survive silicon and software scrutiny is the open question — but the architectural claim itself is the news for overseas infra teams watching China’s HBM-alternative experiments.

What DF1000 is

Press and company framing call DF1000 a software-defined near-memory 3D AI chip. Mechanically, that means:

The “software-defined” half is positioning: soft-defined compute plus 3D near-memory as China’s “third path” versus pure ASIC and GPGPU-shaped designs. Three use cases are cited in coverage: LLM training, AFD-style distributed inference, and single-node LLM inference.

Can you use it today?

Honest answer: no retail SKU, no public cloud listing found, no Digi-Key board, no open weights, nothing overseas labs can buy on Amazon this afternoon. This is architecture + cluster-delivery news from Chinese press (AI123, 至顶科技, 界面新闻, and July securities-journal coverage of the chip launch). Treat it as a supply-chain / HBM-alternative watch item for infra architects — not a purchase order. If a cloud or OEM listing appears later, re-check; as of this writing the deliverable is a delivered domestic cluster story, not an exportable product page.

DF1000 product stack: OAM2.0 card, QY100 8-card server, TY64 64-card supernode, HS128 128-card cluster, bonded wafer detail and DF2000/DF3000 roadmap

Product ladder 巅峯 DF1000 → 擎元 QY100 → 拓域 TY64 → 慧算 HS128, plus bonded-wafer detail and DF2000/DF3000 roadmap*. Asterisks mark vendor/press figures.

Cluster delivery: 128 cards, 72 hours*

Chinese coverage around the Shanghai 国芯展 says the first domestically fully-sourced large-scale training cluster based on DF1000 has been delivered. The headline configuration is the 慧算 HS128 — 128 DF1000 cards — with claimed continuous training for 72 hours without interrupt*. Workloads named in press include Llama2-70B and JYXC-130B / Step 130B-class ~100B-parameter training tasks*, with model adaptation reportedly taking about 7 days* for a 100B-class model. A thousand-card cluster is framed as planned by end-2026*. None of these are independent audits; they are company and trade-press claims.

Product stack

Roadmap slides in coverage point to DF2000 ~2026 Q4 (≈2×*) and DF3000 ~2027 Q4 (≈2× again*). Full domestic supply chain is claimed end-to-end: design → 14nm wafer → 3D packaging → test*.

Company and timeline

东方算芯 was founded May 2024 in Shanghai Zhangjiang. CEO/chairman Wei Shaojun (魏少军) brings a well-known Tsinghua / Chinese semiconductor-policy pedigree — relevant context for overseas readers who track which new AI-chip houses have establishment weight behind them. The DF1000 chip was announced July 13, 2026 (界面新闻 and securities-journal coverage); cluster-delivery stories landed at the Shanghai 国芯展 window, with AI123 updating around Oct 30–31, 2026 and 至顶科技 coverage around mid-October.

How this sits next to other ChinaHotTech chip posts

Do not merge DF1000 with earlier ChinaHotTech chip coverage. Rockchip RK1828, XHEART X7 Ultra, Nuclei NI1000, SpacemiT V100, Infenisis Taiyi, Tianmou, StartLux, Amlogic, and similar posts are different product classes — edge/auto SoCs, RISC-V IP, server CPUs, or developer boards. DF1000 is a near-memory training accelerator + domestic cluster delivery story: HBM-alternative packaging and supply-chain completeness, not a hobbyist board or open IP core.

Caveats

Bottom line: DF1000 is Dongfang Suanxin’s public bet that 3D near-memory bonding on 14nm can substitute for scarce HBM in domestic training clusters — and Chinese press now says a 128-card system has been delivered and run for 72 continuous hours*. Overseas engineers should read it as architecture and supply-chain signal, keep asterisks on every number, and not expect a shopping-cart SKU.

Sources