Lightning-AI/pytorch-lightning · error · ValueError

precision set through both strategy class and plugins, choos

Error message

precision set through both strategy class and plugins, choose one

What it means

A passed-in Strategy instance already has a precision plugin (`_precision` is set), and a Precision plugin was also supplied via `plugins` (or precision resolution already produced one). Fabric refuses to silently pick one.

Source

Thrown at src/lightning/fabric/connector.py:270

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

        self._precision_input = "32-true" if precision_input is None else precision_input

        # 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 isinstance(self._strategy_flag, Strategy):
            if self._strategy_flag._accelerator:
                if self._accelerator_flag != "auto":
                    raise ValueError("accelerator set through both strategy class and accelerator flag, choose one")
                self._accelerator_flag = self._strategy_flag._accelerator
            if self._strategy_flag._precision:
                # [RFC] handle precision plugin set up conflict?
                if self._precision_instance:
                    raise ValueError("precision set through both strategy class and plugins, choose one")
                self._precision_instance = self._strategy_flag._precision
            if self._strategy_flag._checkpoint_io:
                if self.checkpoint_io:
                    raise ValueError("checkpoint_io set through both strategy class and plugins, choose one")
                self.checkpoint_io = self._strategy_flag._checkpoint_io
            if getattr(self._strategy_flag, "cluster_environment", None):
                if self._cluster_environment_flag:
                    raise ValueError("cluster_environment set through both strategy class and plugins, choose one")
                self._cluster_environment_flag = getattr(self._strategy_flag, "cluster_environment")

            if hasattr(self._strategy_flag, "parallel_devices") and self._strategy_flag.parallel_devices:
                if self._strategy_flag.parallel_devices[0].type == "cpu":
                    if self._accelerator_flag and self._accelerator_flag not in ("auto", "cpu"):
                        raise ValueError(
                            f"CPU parallel_devices set through {self._strategy_flag.__class__.__name__} class,"
                            f" but accelerator set to {self._accelerator_flag}, please choose one device type"
                        )
                    self._accelerator_flag = "cpu"

View on GitHub (pinned to 9fed5c27d2)

Solutions

1. Remove the Precision plugin from `plugins` and keep precision inside the strategy
2. Or construct the strategy without precision= and pass the Precision plugin via `plugins`

Example fix

# before
strategy = DDPStrategy(precision=MixedPrecision("16-mixed"))
fabric = Fabric(strategy=strategy, plugins=[MixedPrecision("16-mixed")])

# after
strategy = DDPStrategy(precision=MixedPrecision("16-mixed"))
fabric = Fabric(strategy=strategy)

from lightning.fabric.plugins import Precision

if isinstance(strategy, Strategy) and strategy._precision and any(isinstance(p, Precision) for p in plugins):
    raise SystemExit("Precision defined both in strategy and plugins; keep one")

When it happens

Trigger: strategy = DDPStrategy(precision=MixedPrecision(...)); Fabric(strategy=strategy, plugins=[MixedPrecision(...)]) — strategy._precision non-None while self._precision_instance is set.

Common situations: Preconfigured strategy objects (DeepSpeedStrategy often bundles a precision plugin) combined with a plugins list; migrating Trainer setups where precision lived in the strategy; factory-created strategies reused with new plugin configs.

Related errors

• Received both `precision={precision_input}` and `plugins={se
• checkpoint_io set through both strategy class and plugins, c
• cluster_environment set through both strategy class and plug
• accelerator set through both strategy class and accelerator
• Received multiple values for {', '.join(duplicated_plugin_ke

AI-assisted analysis of Lightning-AI/pytorch-lightning@9fed5c27d2 (2026-08-28). Data as JSON: /api/errors/b71a8c67c18eafba. Report an issue: GitHub.