## Does China's Newest Neutral-Atom QPU Actually Fit in a Data-Center Rack?

A Chinese Academy of Sciences spinout just made a direct challenge to cryogenic quantum infrastructure — but without publishing a single benchmark.

Zhongqi Wuliang (众擎无量) formally unveiled **Qinghe No. 1** (清河一号) at the World Artificial Intelligence Conference (WAIC 2026) in Shanghai on July 20, 2026. The second-generation [neutral atom qubit](https://quantumintel.tech/glossary/neutral-atom-qubit) system is engineered to fit inside a standard data-center server rack — eliminating the [dilution refrigerator](https://quantumintel.tech/glossary/dilution-refrigerator) infrastructure that superconducting systems from [IBM Quantum](https://quantumintel.tech/companies/ibm) and [Google Quantum AI](https://quantumintel.tech/companies/google-quantum-ai) require to maintain operation at approximately 15 millikelvin. Founded by Lu Xudong and spun out of the CAS Shanghai Institute of Optics and Fine Mechanics (CAS SIOM), the company claims the rack form factor enables cloud providers to co-locate quantum hardware alongside classical servers. Hardware integration was enabled in part by supply chain partner Xuanxiang Technology, which has commercialized a metasurface chip for generating large-scale optical tweezer arrays. The critical caveat: Zhongqi Wuliang has not yet published peer-reviewed qubit counts, two-qubit [gate fidelity](https://quantumintel.tech/glossary/gate-fidelity) figures, or [coherence time](https://quantumintel.tech/glossary/coherence-time) data for Qinghe No. 1. Until independent benchmarks appear, enterprise buyers and investors should treat this as a form-factor proof-of-concept, not a validated compute platform.

---

## What Qinghe No. 1 Actually Is

The core engineering proposition is about packaging, not necessarily performance. Conventional superconducting quantum processors require dilution refrigerators — large, expensive, vibration-sensitive cryogenic systems that cool physical circuits to roughly 15 millikelvin. This infrastructure is physically incompatible with standard enterprise data centers: the footprint is large, the energy overhead is significant, and the vibration isolation requirements are exacting.

Neutral-atom processors sidestep this entirely. Atoms (typically rubidium or cesium) are trapped using laser-generated optical tweezers and manipulated with precisely controlled laser pulses. The vacuum chamber operates at room temperature relative to the superconducting baseline — the laser and optical control systems are the dominant hardware complexity, not cryogenic refrigeration.

Zhongqi Wuliang's stated innovation is integrating the vacuum chamber, control optics, and laser delivery systems into a single-rack chassis. Supply chain partner Xuanxiang Technology's metasurface chip — described as capable of generating large-scale optical tweezer arrays — is cited as a key enabler for simplifying optical alignment, which has historically been a significant barrier to neutral-atom miniaturization.

This is a genuinely non-trivial engineering challenge. Optical alignment at the precision required for quantum gate operations is sensitive to thermal drift, mechanical vibration, and component aging. Packaging all of this reliably into a rack-mountable unit, at production scale, remains an open problem that no company has fully solved at commercial volume — including Western peers like [QuEra Computing](https://quantumintel.tech/companies/quera-computing) and [Pasqal](https://quantumintel.tech/companies/pasqal).

---

## China's Neutral-Atom Cluster Is Moving Quickly

Qinghe No. 1 is not an isolated development. According to the source reporting, it is the third commercially positioned neutral-atom system to emerge from China's regional quantum hardware cluster in less than a year. The source does not name the other two systems explicitly, but the pace of iteration is notable regardless of individual system specifications.

This cadence is directly supported by municipal industrial policy. Shanghai has designated neutral-atom technology as a core element of its quantum roadmap, establishing dedicated innovation zones — including the Xuhui Cultivation Zone and Zhangjiang Quantum Bay — backed by state research funds and municipal subsidies. This kind of coordinated infrastructure investment, linking academic spinouts like Zhongqi Wuliang to both supply chain partners and guaranteed soft landing zones, mirrors the playbook that accelerated China's photovoltaic and semiconductor capacity in prior technology cycles.

For Western quantum hardware companies, the relevant signal is not Qinghe No. 1's current performance — which is unverified — but the institutional infrastructure being built around it. State-backed innovation zones lower the survival threshold for early-stage hardware companies, allowing more shots on goal during the critical period before any neutral-atom system achieves fault-tolerant operation.

---

## The Benchmark Gap Is the Real Story

Technical buyers evaluating Qinghe No. 1 for near-term deployment face a fundamental problem: there is no published data to evaluate. The source explicitly notes the absence of peer-reviewed metrics for qubit count, two-qubit gate fidelities, or coherence times. Without these figures, comparison against established global neutral-atom benchmarks is impossible.

This is not unusual for a freshly unveiled system — first public demonstrations frequently precede full technical disclosure. But the gap matters more in the neutral-atom modality than in superconducting, where the IBM and Google ecosystems have established dense public benchmarking cultures including quantum volume, [CLOPS](https://quantumintel.tech/glossary/clops), and layer fidelity metrics. Neutral-atom players globally, including QuEra and Pasqal, have published detailed gate fidelity and coherence data to support enterprise evaluation. Zhongqi Wuliang will need to meet that standard to be taken seriously outside of China's domestic procurement ecosystem.

The form factor story is compelling on paper. But a rack-mounted QPU with unknown fidelity is not a deployable enterprise asset — it is a development-stage prototype with good industrial design.

---

## Google's Parallel Move Adds Context

The announcement coincides with a broader global shift in neutral-atom investment. Earlier in 2026, Google Quantum AI launched a dedicated neutral-atom research program in Boulder, Colorado, running in parallel to its superconducting work. Google's stated rationale — reconfigurable qubit connectivity and scalability — echoes exactly the architectural arguments Zhongqi Wuliang is implicitly making with a rack-deployable system.

The convergence of a major Western incumbent and multiple Chinese startups on the same modality, within the same 12-month window, is a stronger signal of neutral atoms' medium-term trajectory than any single product announcement. The question is no longer whether neutral atoms can compete with superconducting at a physics level — it is which organizations will solve the engineering and reliability challenges at production scale first.

---

## Key Takeaways

- **Qinghe No. 1** is a second-generation rack-mounted neutral-atom QPU from Chinese CAS SIOM spinout Zhongqi Wuliang, unveiled at WAIC 2026 in Shanghai on July 20, 2026.
- The system eliminates dilution refrigerator infrastructure by operating neutral atoms at non-cryogenic conditions, targeting direct deployment in enterprise data centers.
- Supply chain partner Xuanxiang Technology contributed a metasurface chip for large-scale optical tweezer array generation, addressing a key miniaturization bottleneck.
- **No peer-reviewed benchmarks have been published** — qubit count, two-qubit gate fidelity, and coherence times are all unverified as of this writing.
- Qinghe No. 1 is reportedly the third commercially positioned neutral-atom system from China's hardware cluster in under a year, supported by Shanghai municipal quantum innovation zones including Zhangjiang Quantum Bay.
- Google Quantum AI's parallel launch of a neutral-atom research program in Boulder in early 2026 signals broad Western validation of the modality's long-term scalability case.
- Enterprise buyers should wait for independent benchmarks before evaluating Qinghe No. 1 against QuEra, Pasqal, or other established neutral-atom platforms.

---

## Frequently Asked Questions

**What is Qinghe No. 1 and who makes it?**
Qinghe No. 1 (清河一号) is a neutral-atom quantum computer built by Zhongqi Wuliang (众擎无量), a Chinese deep-tech startup founded by Lu Xudong and spun out from the Chinese Academy of Sciences' Shanghai Institute of Optics and Fine Mechanics. It was publicly unveiled at WAIC 2026 in Shanghai on July 20, 2026.

**Why does a server-rack form factor matter for quantum computing?**
Standard superconducting quantum processors require dilution refrigerators operating at approximately 15 millikelvin — infrastructure that is physically incompatible with conventional enterprise data centers due to size, energy, and vibration isolation requirements. Neutral-atom systems do not require cryogenic cooling at this scale, making rack-mounted deployment alongside classical servers architecturally feasible.

**What are the published performance specs for Qinghe No. 1?**
As of the WAIC 2026 announcement, Zhongqi Wuliang has not published peer-reviewed metrics including qubit count, two-qubit gate fidelity, or coherence times. Independent benchmarking against established neutral-atom systems has not yet occurred.

**How does China's neutral-atom program compare to Western efforts?**
China's regional neutral-atom hardware cluster has produced at least three commercially positioned systems in under a year, supported by Shanghai municipal policy infrastructure including Zhangjiang Quantum Bay and the Xuhui Cultivation Zone. Western competitors including QuEra Computing and Pasqal have more established public benchmark records. Google Quantum AI launched a dedicated neutral-atom research program in Boulder, Colorado in early 2026.

**What should enterprise buyers watch for before evaluating Qinghe No. 1?**
The critical missing data points are published qubit counts, two-qubit gate fidelities, and coherence times — preferably from peer-reviewed or independently verified sources. The rack form factor is an engineering concept until these figures confirm that gate performance meets operational thresholds for meaningful computation.