CUDA-Q Remote Simulation Runtime REST API (1.0.0)

Download OpenAPI specification:Download

Create a quantum simulation job

Requests execution (simulation) of quantun kernels on the server. The server employs long polling, hence it will keep the request open until the simulation completed.

Request Body schema: application/json

version required	integer <int64> Version number of the payload. Each REST server version only supports a specific payload version.
clientVersion	string Version description of the CUDA-Q client generating the request.
entryPoint required	string Entry point kernel identifier as a fully-qualified name.
simulator	string Name of the backend simulator to be used.
required	object (RequestExecutionContext)
code required	string <binary> Base64 encoded CUDA-Q kernel IR (Intermediate Representation).
args	string <binary> Base64 encoded kernel arguments.
format required	string Enum: "MLIR" "LLVM" Format of the code IR.
seed	integer <int64> Random seed value for the simulation.
passes	Array of strings List of IR passes to be applied on the server.

Responses

Response Schema: /

required

object (ResultExecutionContext)

Request samples

Payload

Content type

application/json

{"version": 1,
"clientVersion": "CUDA-Q Version 0.7.0 (https://github.com/NVIDIA/cuda-quantum 8650fb9d79e7e780ea95d9506a513162daf84b80)",
"entryPoint": "ns::ghz",
"simulator": "nvidia-mgpu",
"executionContext": {"name": "sample",
"shots": 1000,
"hasConditionalsOnMeasureResults": "false",
"registerNames": ["string"
],
"spin": {"num_qubits": 4,
"data": [0.1
]
}
},
"code": "string",
"args": "string",
"format": "MLIR",
"seed": 123,
"passes": ["inline",
"canonicalize",
"quake-to-qir"
]
}