langchain-ai/langgraph · error · TypeError

StreamChannel is bound to async mode — sync interleave()…

Error message

StreamChannel {name!r} is bound to async mode — sync interleave() cannot consume async channels.

What it means

Raised by the synchronous interleave() when a requested StreamChannel was bound to async mode. A sync cursor cannot pump an async-bound channel, so mixing sync consumption with async channels is rejected with a TypeError. Use the async interleave variant or bind the channel to sync mode.

Solutions

1. Use the async interleave() (under astream) for async-bound channels
2. Bind the channel in sync mode by running sync stream()

When it happens

Trigger: Thrown at libs/langgraph/langgraph/stream/run_stream.py:270 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of langchain-ai/langgraph@38031739e5 (2026-08-26). Data as JSON: /api/errors/d3efa5c4440e7b04. Report an issue: GitHub.

Appendix: source

Thrown at libs/langgraph/langgraph/stream/run_stream.py:270

        """
        from langgraph.stream.stream_channel import StreamChannel

        channels: dict[str, StreamChannel[Any]] = {}
        try:
            for name in names:
                ch = self.extensions[name]
                if not isinstance(ch, StreamChannel):
                    raise TypeError(
                        f"interleave() requires StreamChannel projections, "
                        f"got {type(ch).__name__} for {name!r}"
                    )
                if ch._is_async is None:
                    raise TypeError(
                        f"StreamChannel {name!r} has not been bound yet. "
                        "Register the transformer with a StreamMux first."
                    )
                if ch._is_async:
                    raise TypeError(
                        f"StreamChannel {name!r} is bound to async mode — "
                        "sync interleave() cannot consume async channels."
                    )
                if ch._subscribed:
                    raise RuntimeError(
                        f"StreamChannel {name!r} already has a subscriber; "
                        "use .tee(n) for fan-out."
                    )
                ch._subscribed = True
                channels[name] = ch

            done: set[str] = set()

            while len(done) < len(channels):
                best: tuple[int, str] | None = None
                for name, ch in channels.items():
                    if name in done:
                        continue

View on GitHub (pinned to 38031739e5)