2noise/ChatTTS · error · ValueError

GPU memory utilization must be less than 1.0. Got {self.gpu_

Error message

GPU memory utilization must be less than 1.0. Got {self.gpu_memory_utilization}.

What it means

Error "GPU memory utilization must be less than 1.0. Got {self.gpu_memory_utilization}." thrown in 2noise/ChatTTS.

Source

Thrown at ChatTTS/model/velocity/configs.py:317

        self,
        block_size: int,
        gpu_memory_utilization: float,
        swap_space: int,
        sliding_window: Optional[int] = None,
    ) -> None:
        self.block_size = block_size
        self.gpu_memory_utilization = gpu_memory_utilization
        self.swap_space_bytes = swap_space * _GB
        self.sliding_window = sliding_window
        self._verify_args()

        # Will be set after profiling.
        self.num_gpu_blocks = None
        self.num_cpu_blocks = None

    def _verify_args(self) -> None:
        if self.gpu_memory_utilization > 1.0:
            raise ValueError(
                "GPU memory utilization must be less than 1.0. Got "
                f"{self.gpu_memory_utilization}."
            )

    def verify_with_parallel_config(
        self,
        parallel_config: "ParallelConfig",
    ) -> None:
        total_cpu_memory = get_cpu_memory()
        # FIXME(woosuk): Here, it is assumed that the GPUs in a tensor parallel
        # group are in the same node. However, the GPUs may span multiple nodes.
        num_gpus_per_node = parallel_config.tensor_parallel_size
        cpu_memory_usage = self.swap_space_bytes * num_gpus_per_node

        msg = (
            f"{cpu_memory_usage / _GB:.2f} GiB out of "
            f"the {total_cpu_memory / _GB:.2f} GiB total CPU memory is "
            "allocated for the swap space."

View on GitHub (pinned to 77b89ee281)

When it happens

Trigger: Thrown at ChatTTS/model/velocity/configs.py:317 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of 2noise/ChatTTS@77b89ee281 (2026-08-26). Data as JSON: /api/errors/abc9bc58cfd3917a. Report an issue: GitHub.