openai/openai-python · error · ValueError

Expected a non-empty value for `call_id` but received {call_

Error message

Expected a non-empty value for `call_id` but received {call_id!r}

What it means

Raised by RealtimeCalls.sync accept() when the `call_id` argument is falsy (empty string or None). The SDK validates path parameters before making the HTTP request because an empty call_id would produce a malformed URL (/realtime/calls//accept). It is a client-side guard, not a server response.

Source

Thrown at src/openai/resources/realtime/calls.py:265

              cache), since messages are dropped from the beginning of the context. However,
              clients can also configure truncation to retain messages up to a fraction of the
              maximum context size, which will reduce the need for future truncations and thus
              improve the cache rate.

              Truncation can be disabled entirely, which means the server will never truncate
              but would instead return an error if the conversation exceeds the model's input
              token limit.

          extra_headers: Send extra headers

          extra_query: Add additional query parameters to the request

          extra_body: Add additional JSON properties to the request

          timeout: Override the client-level default timeout for this request, in seconds
        """
        if not call_id:
            raise ValueError(f"Expected a non-empty value for `call_id` but received {call_id!r}")
        extra_headers = {"Accept": "*/*", **(extra_headers or {})}
        return self._post(
            path_template("/realtime/calls/{call_id}/accept", call_id=call_id),
            body=maybe_transform(
                {
                    "type": type,
                    "audio": audio,
                    "include": include,
                    "instructions": instructions,
                    "max_output_tokens": max_output_tokens,
                    "model": model,
                    "output_modalities": output_modalities,
                    "parallel_tool_calls": parallel_tool_calls,
                    "prompt": prompt,
                    "reasoning": reasoning,
                    "tool_choice": tool_choice,
                    "tools": tools,
                    "tracing": tracing,

View on GitHub (pinned to 9917c6e28e)

Solutions

Ensure call_id comes from a valid source, e.g. the `call` object's id from a realtime event or client.realtime.calls.create()
Check for a non-empty value before calling accept(): if not call_id: raise/log
If call_id is Optional in your flow, guard the whole call behind `if call_id is not None`

Example fix

// before
client.realtime.calls.accept(call_id=call_id or "")
# after
if not call_id:
    raise ValueError("call_id missing; was the call created?")
client.realtime.calls.accept(call_id=call_id)

Defensive patterns

Strategy: validation

Validate before calling

if not isinstance(call_id, str) or not call_id.strip():
    raise ValueError(f"invalid call_id: {call_id!r}")
client.realtime.calls.accept(call_id=call_id)

Type guard

def is_valid_call_id(v: object) -> bool:
    return isinstance(v, str) and bool(v.strip())

Try / catch

try:
    client.realtime.calls.accept(call_id=call_id)
except ValueError as e:
    logger.error("bad call_id: %s", e)

Prevention

Always source call_id from the call resource or call events
Never default call_id to an empty string
Log ids when calls are created

When it happens

Trigger: Calling client.realtime.calls.accept(call_id="") or call_id=None (or a variable that was never populated, e.g. an empty field from a call.created event payload).

Common situations: Storing call_id from an event where the field was absent, typo'd variable, defaulting call_id to "" instead of omit, or passing a falsy sentinel before the call object was created.

Understand the failure class

Background: "Must be a positive integer", "Invalid value", "Unsupported": the invalid-argument-value error family, when a library rejects the value you pass — this error's family across 35 libraries.

Related errors

AI-assisted analysis of openai/openai-python@9917c6e28e (2026-08-28). Data as JSON: /api/errors/bb1ac3d4d156e9a3. Report an issue: GitHub.