Python SDK
Submit, wait, read the counts and fetch the evidence from a program of your own.
pip install softqcos-sdk
Development infrastructure
QCOS meets you in Python, in a terminal or through an assistant. The first run is on a CPU simulator, so you can learn the whole flow before you think about a machine.
Submit, wait, read the counts and fetch the evidence from a program of your own.
pip install softqcos-sdk
The same jobs from a terminal or a script, down to offline verification of the evidence.
softqcos doctor
The official QCOS server for assistants and agents that speak the Model Context Protocol.
softqcos-mcp
The contract underneath the three of them, for anything else that speaks HTTP.
/api/v1/runtime
from softqcos_sdk import QCOSClient
qcos = QCOSClient(api_key="your-api-key")
# See what the job will take, without running it
estimate = qcos.jobs.estimate(
qasm=bell_state,
backend_id="simulator",
shots=1000,
)
job = qcos.jobs.submit(
qasm=bell_state,
backend_id="simulator",
shots=1000,
)
result = qcos.jobs.wait(job["job_id"])
print(result["result"]["counts"])
# Every run records an evidence bundle
print(qcos.jobs.artifacts(job["job_id"]))$ softqcos doctor
$ softqcos jobs submit \
--circuit circuits/bell-state.qasm \
--backend simulator \
--shots 1000
$ softqcos jobs status <job-id>
$ softqcos jobs results <job-id>
$ softqcos jobs evidence <job-id> \
--output evidence.json
$ softqcos evidence verify evidence.json --offlineWhat we are building next
Call ordinary C and Python functions from a quantum program, and see what they will cost in qubits first. HiQ is a design today; there is no compiler to download yet.
Write, estimate and run from the editor you already use, without leaving your file.
Try a job against a software stand-in for the machine you chose, before you spend anything on it.
The film is an original render.