Hands-On: Origin Wukong (本源悟空)
Origin Wukong is China's home-grown superconducting quantum computer, open to the world over a public cloud. This page shows how to prepare a kernel and submit it through Origin's own SDK; the qalgora-Q reference build itself runs locally and does not submit natively. The latest Wukong-180 carries 180 compute qubits.
1 · Get cloud access
- Register on the Origin Quantum cloud and obtain an access token.
- Export it:
export QALGORA_ORIGIN_TOKEN="..."
2 · Validate locally, then submit (adapter workflow · planned)
import qalgora
@qalgora.kernel
def bell():
q = qalgora.qvector(2)
h(q[0]); x.ctrl(q[0], q[1]); mz(q)
qalgora.set_target("qpp-cpu") # reference implementation, local validation
print(qalgora.sample(bell, shots_count=1000))# qalgora.set_target("origin", machine="wukong") # planned adapter — needs Origin SDK + credentials
# job = qalgora.sample_async(bell, shots_count=4000)
# print(job.status())
# print(job.get()) # adapter future; blocks until the remote job finishesjob.get()3 · Gate-based — but not zero-adaptation
Wukong runs standard gate-based kernels, so a qalgora-Q program targets it by switching the target name — but submission is not zero-adaptation: the adapter still performs gate-set translation, routing to the device's coupling map, and shot/queue handling. Coupler qubits boost connectivity.
4 · Why Origin for learning
- Open public cloud — accessible worldwide, good for first hardware runs.
- Self-contained stack — operating system and SDK provided by Origin.
- Hundreds of thousands of jobs have already run through it (tens of millions of visits), so the queue and tooling are mature.
qpp-cpu first to save queue time.
实战：Origin Wukong（本源悟空）
本源悟空是中国自主研发的超导量子计算机，通过公有云向全球开放。本页演示如何准备内核并经本源自己的 SDK 提交；qalgora-Q 参考实现本身在本地运行，不提供原生真机提交。最新的 Wukong-180 搭载 180 个计算量子比特。
1 · 获取云端访问权限
- 在本源量子云平台注册账号并获取访问令牌。
- 导出令牌：
export QALGORA_ORIGIN_TOKEN="..."
2 · 本地验证，然后提交（适配器工作流·规划中）
import qalgora
@qalgora.kernel
def bell():
q = qalgora.qvector(2)
h(q[0]); x.ctrl(q[0], q[1]); mz(q)
qalgora.set_target("qpp-cpu") # 参考实现本地验证
print(qalgora.sample(bell, shots_count=1000))# qalgora.set_target("origin", machine="wukong") # 规划中 / 适配示意：需本源 SDK 与凭据
# job = qalgora.sample_async(bell, shots_count=4000)
# print(job.status())
# print(job.get()) # 适配器 future；阻塞直到远程任务完成job.get()3 · 基于门的架构——但并非零适配
本源悟空运行标准的基于门的内核，因此 qalgora-Q 程序只需切换目标名称即可指向它——但提交并非零适配：适配器仍需进行门集编译、按设备耦合图布线，以及采样次数/排队处理。耦合量子比特提升了连通性。
4 · 为何用本源来学
- 开放公有云——面向全球开放，适合首次真机运行。
- 自成一体的技术栈——操作系统与 SDK 均由本源提供。
- 已有数十万个任务在此运行（数千万次访问），排队机制与工具链均已成熟。