Skip to main content
Set stream: true on POST /v1/responses and the reply arrives as standard OpenAI Responses response.* SSE events — every frame is event: <type> + data: <json>, with a monotonic sequence_number. Official SDK stream loops consume it as-is. 
from openai import OpenAI

client = OpenAI(api_key="mb_live_...", base_url="https://test-mirobody-api.thetahealth.ai/v1")

with client.responses.stream(
    model="mirobody-flash",
    input="How is my fasting glucose trending?",
    user="alice",
) as stream:
    for event in stream:
        if event.type == "response.output_text.delta":
            print(event.delta, end="", flush=True)

​
Event table

EventMeaning
response.createdThe run started — carries a response snapshot with status: "in_progress".
response.in_progressFollows immediately (OpenAI parity).
response.output_item.addedA new output item opened (reasoning, message, or function_call) — carries output_index + the in-progress item.
response.reasoning_summary_part.added / .doneA reasoning summary part opened / closed.
response.reasoning_summary_text.delta / .doneReasoning (deep-thinking) text fragments, then the full text.
response.content_part.added / .doneAn answer content part opened / closed.
response.output_text.delta / .doneAnswer text fragments, then the full text — the channel most clients render.
response.function_call_arguments.delta / .doneA client-tool handoff’s arguments (see Function calling).
response.output_item.doneThe open item completed — carries the finished item.
response.mirobody_tool_callMirobody extension, side-channel — a built-in server tool ran. See below.
response.completedTerminal: carries the full final response object (output, usage, tool_steps, health_records, citations).
response.failedTerminal error: carries a response snapshot with status: "failed" and an error object.
Items stream strictly one at a time: reasoning first (when the tier emits it), then the message, then any function_call handoffs — output_index increments per item, and the indexes match the final output array on response.completed.

​
Server-tool side-channel

Built-in server tools (data search, literature, …) are not output items — official SDKs would mis-parse unknown item types, so their trace rides a dedicated event that standard stream loops safely ignore: 
event: response.mirobody_tool_call
data: {"type":"response.mirobody_tool_call","sequence_number":7,
       "tool_step":{"id":"mtc_0","call_id":"call_1a2b...","name":"query_health_data",
                    "arguments":"{\"query\": \"fasting glucose last 90 days\"}"}}
It fires when the tool is called; it does not open an output item or advance output_index. Full results (tool_steps[].result) are on the final response object in response.completed. Render it if you want an activity feed (“Searching your records…”); skip it and nothing breaks.

​
Failure shape

On an upstream error the stream ends with response.failed (not a broken pipe): 
event: response.failed
data: {"type":"response.failed","sequence_number":12,
       "response":{"id":"resp_...","status":"failed",
                   "error":{"type":"upstream_error","message":"..."}}}
For the Answers API’s simpler chat.completion.chunk streaming, see Answers API → Streaming.