Lightning-AI/pytorch-lightning · error · ValueError

Received both `precision={precision_flag}` and `plugins={sel

Error message

Received both `precision={precision_flag}` and `plugins={self._precision_plugin_flag}`. Choose one.

What it means

Precision can be configured either via the convenience Trainer(precision=...) flag or via a Precision plugin in plugins=[...], but not both. Note this specific check raises ValueError (not MisconfigurationException).

Source

Thrown at src/lightning/pytorch/trainer/connectors/accelerator_connector.py:260

                            " plugin, but this is not allowed. Choose one or the other."
                        )
                    self._layer_sync = plugin
                    plugins_flags_types[TorchSyncBatchNorm.__name__] += 1
                else:
                    raise MisconfigurationException(
                        f"Found invalid type for plugin {plugin}. Expected one of: Precision, "
                        "CheckpointIO, ClusterEnvironment, or LayerSync."
                    )

            duplicated_plugin_key = [k for k, v in plugins_flags_types.items() if v > 1]
            if duplicated_plugin_key:
                raise MisconfigurationException(
                    f"Received multiple values for {', '.join(duplicated_plugin_key)} flags in `plugins`."
                    " Expected one value for each type at most."
                )

            if plugins_flags_types.get(Precision.__name__) and precision_flag is not None:
                raise ValueError(
                    f"Received both `precision={precision_flag}` and `plugins={self._precision_plugin_flag}`."
                    f" Choose one."
                )

        self._precision_flag = "32-true" if precision_flag is None else precision_flag

        # handle the case when the user passes in a strategy instance which has an accelerator, precision,
        # checkpoint io or cluster env set up
        # TODO: improve the error messages below
        if self._strategy_flag and isinstance(self._strategy_flag, Strategy):
            if self._strategy_flag._accelerator:
                if self._accelerator_flag != "auto":
                    raise MisconfigurationException(
                        "accelerator set through both strategy class and accelerator flag, choose one"
                    )
                self._accelerator_flag = self._strategy_flag._accelerator
            if self._strategy_flag._precision_plugin:
                # [RFC] handle precision plugin set up conflict?

View on GitHub (pinned to 9fed5c27d2)

Solutions

1. Remove the precision= flag and keep only the plugin (or vice versa)
2. If the plugin was added by shared helper code, parameterize it so it is only added when no precision flag is set

Example fix

# before
trainer = Trainer(precision="bf16-mixed", plugins=[MyPrecisionPlugin()])
# after
trainer = Trainer(plugins=[MyPrecisionPlugin()])

from lightning.pytorch.plugins import Precision
has_precision_plugin = any(isinstance(p, Precision) for p in plugins)
if has_precision_plugin:
    precision = None  # let the plugin win
trainer = Trainer(precision=precision, plugins=plugins)

def precision_conflict(precision, plugins) -> bool:
    from lightning.pytorch.plugins import Precision
    return precision is not None and any(isinstance(p, Precision) for p in (plugins or []))

try:
    trainer = Trainer(precision=precision, plugins=plugins)
except ValueError as e:
    if "Received both" in str(e):
        trainer = Trainer(plugins=plugins)
    else:
        raise

When it happens

Trigger: Trainer(precision='16-mixed', plugins=[MixedPrecision('16-mixed')]) or any combination where a Precision instance is in plugins and precision flag is not None.

Common situations: Upgrading code that sets precision='16-mixed' while a tutorial/dependency also injects a precision plugin; refactors that add a custom precision plugin without removing the old precision flag.

Related errors

• precision set through both strategy class and plugins, choos
• Passed `{type(self).__name__}(precision={precision!r})`. Pre
• `precision='bf16-mixed'` does not use a scaler, found {scale
• `precision={precision!r})` is not supported in DeepSpeed. `p
• `precision={precision!r})` is not supported in FSDP. `precis

AI-assisted analysis of Lightning-AI/pytorch-lightning@9fed5c27d2 (2026-08-28). Data as JSON: /api/errors/9a116601e5de808f. Report an issue: GitHub.