langchain-ai/langchain · error · ValueError

config must be a list of the same length as inputs, but got

Error message

config must be a list of the same length as inputs, but got {len(config)} configs for {length} inputs

What it means

When a `config` for a batch operation is given as a sequence, `get_config_list` requires `len(config) == length`, where `length` is the number of inputs being processed. A mismatch means some inputs would have no config (or configs would be silently dropped), so `ValueError` is raised reporting both counts.

Source

Thrown at libs/core/langchain_core/runnables/config.py:337

        config: The config or list of configs.
        length: The length of the list.

    Returns:
        The list of configs.

    Raises:
        ValueError: If the length of the list is not equal to the length of the inputs.

    """
    if length < 0:
        msg = f"length must be >= 0, but got {length}"
        raise ValueError(msg)
    if isinstance(config, Sequence) and len(config) != length:
        msg = (
            f"config must be a list of the same length as inputs, "
            f"but got {len(config)} configs for {length} inputs"
        )
        raise ValueError(msg)

    if isinstance(config, Sequence):
        return list(map(ensure_config, config))
    if length > 1 and isinstance(config, dict) and config.get("run_id") is not None:
        warnings.warn(
            "Provided run_id be used only for the first element of the batch.",
            category=RuntimeWarning,
            stacklevel=3,
        )
        subsequent = cast(
            "RunnableConfig", {k: v for k, v in config.items() if k != "run_id"}
        )
        return [
            ensure_config(subsequent) if i else ensure_config(config)
            for i in range(length)
        ]
    return [ensure_config(config) for i in range(length)]

View on GitHub (pinned to e32fa9a52e)

Solutions

Make the config list match the inputs one-to-one: build it with a list comprehension over the inputs.
If all inputs share one config, pass a single dict instead of a list: `.batch(inputs, config=shared_cfg)`.
When sharding inputs, shard the configs identically: `.batch(inputs[i:j], config=configs[i:j])`.

Example fix

# before
results = chain.batch(inputs, config=[cfg1, cfg2])  # len(inputs) == 3

# after
results = chain.batch(inputs, config=[make_cfg(i) for i in range(len(inputs))])

Defensive patterns

Strategy: validation

Validate before calling

if isinstance(config, list):
    assert len(config) == len(inputs), f"{len(config)} configs vs {len(inputs)} inputs"
chain.batch(inputs, config=config)

Try / catch

try:
    chain.batch(inputs, config=configs)
except ValueError as e:
    if "same length as inputs" in str(e):
        configs = [configs[i % len(configs)] for i in range(len(inputs))]
        chain.batch(inputs, config=configs)
    else:
        raise

Prevention

Build config lists with a comprehension over inputs so lengths always match.
Pass a single shared dict when all items use the same config.
Shard configs together with inputs when parallelizing.

When it happens

Trigger: Calling `.batch(inputs, config=[cfg1, cfg2])` with 3 inputs, or `.batch([x], config=[cfg1, cfg2])`; also custom code that zips inputs with a differently-sized config list, or slices inputs (`inputs[1:]`) while passing the full config list.

Common situations: Reusing a config list built for a previous batch size; parallel workers processing shards of inputs with the unsharded config list; passing `config=` per-item dicts when a single shared `RunnableConfig` dict was intended (and vice versa when a list was intended).

Related errors

AI-assisted analysis of langchain-ai/langchain@e32fa9a52e (2026-08-14). Data as JSON: /api/errors/28cc72b57df10256. Report an issue: GitHub.