Agents

The agents endpoint connects you to Coda’s AI agents, which can write quantum algorithms, generate and optimise circuits, explain concepts, and assist with quantum computing workflows. Responses are streamed back as server-sent events (SSE).

Modes

Example

1
from conductorquantum import ConductorQuantum
2
3
client = ConductorQuantum(token="<your-api-key>")
4
for event in client.coda.agents.run(
5
    messages=[{"role": "user", "content": "Create a 3-qubit GHZ state circuit in Qiskit"}],
6
    mode="build",
7
):
8
    print(event)

The response is streamed. Each event contains a type field (content, tool_call, completed, or error) and associated data.

Conversation threads

Pass a thread_id to continue a previous conversation:

1
{
2
  "messages": [
3
    {"role": "user", "content": "Now add a measurement to all qubits"}
4
  ],
5
  "thread_id": "<thread-id-from-previous-response>"
6
}

The X-Thread-Id response header contains the thread ID for new conversations.

Fast mode

Set "fast": true for lower latency responses using a faster model. This trades some response quality for speed.

Related endpoints

• POST /agents: chat with a Coda AI agent
• GET /agents: list available agent modes

See the Coda API Reference for full request and response schemas.