opendatalab/MinerU · error · ValueError

Unsupported operating system.

Error message

Unsupported operating system.

What it means

Raised inside set_lmdeploy_backend() on the CUDA path when the OS is neither Windows nor Linux as detected by is_windows_environment()/is_linux_environment(). lmdeploy's engine selection (turbomind/pytorch by compute capability) is only implemented for those two platforms.

Source

Thrown at mineru/backend/vlm/utils.py:77

def set_lmdeploy_backend(device_type: str) -> str:
    if device_type.lower() in ["ascend", "maca", "camb"]:
        lmdeploy_backend = "pytorch"
    elif device_type.lower() in ["cuda"]:
        import torch
        if not torch.cuda.is_available():
            raise ValueError("CUDA is not available.")
        if is_windows_environment():
            lmdeploy_backend = "turbomind"
        elif is_linux_environment():
            major, minor = torch.cuda.get_device_capability()
            compute_capability = f"{major}.{minor}"
            if version.parse(compute_capability) >= version.parse("8.0"):
                lmdeploy_backend = "pytorch"
            else:
                lmdeploy_backend = "turbomind"
        else:
            raise ValueError("Unsupported operating system.")
    else:
        raise ValueError(f"Unsupported lmdeploy device type: {device_type}")
    return lmdeploy_backend


def set_default_gpu_memory_utilization() -> float:
    from vllm import __version__ as vllm_version
    device = get_device()
    gpu_memory = get_vram(device)
    default_gpu_memory_utilization = 0.5
    if version.parse(vllm_version) >= version.parse("0.11.0") and gpu_memory <= 8:
        default_gpu_memory_utilization = 0.7

    logger.debug(f"vllm_version: {vllm_version}, gpu_memory: {gpu_memory} GB, default_gpu_memory_utilization: {default_gpu_memory_utilization}")
    return default_gpu_memory_utilization


def set_default_batch_size() -> int:

View on GitHub (pinned to 4fe4bde114)

Solutions

1. Run lmdeploy-engine on Linux or Windows with an NVIDIA GPU.
2. On macOS, switch to the mlx-engine backend (Apple Silicon) or transformers backend.
3. If the platform is genuinely Linux/Windows but detection fails, check platform.system()/env overrides your shell sets.

Example fix

# before
backend = set_lmdeploy_backend("cuda")  # on macOS -> Unsupported operating system.

# after (Apple Silicon Mac)
run_parse(backend="mlx-engine", ...)

import platform

def lmdeploy_cuda_platform_ok() -> bool:
    return platform.system() in ("Windows", "Linux")

import platform

def supports_lmdeploy_cuda() -> bool:
    import torch
    return platform.system() in ("Windows", "Linux") and torch.cuda.is_available()

try:
    backend = set_lmdeploy_backend("cuda")
except ValueError as e:
    if "Unsupported operating system" in str(e):
        backend = "pytorch"  # ascend/maca/camb path, or switch backends entirely
    else:
        raise

When it happens

Trigger: device_type='cuda' on macOS or another Unix where a CUDA device is reported (rare, e.g. legacy Mac CUDA, BSD with NVIDIA drivers) or environment detection misclassifies the platform.

Common situations: Running on macOS with an old CUDA card; esoteric OSes; container reporting a strange platform via platform module.

Related errors

• CUDA is not available.
• Please install lmdeploy to use the lmdeploy-engine backend.
• Unsupported lmdeploy device type: {device_type}
• Unsupported lmdeploy backend: {lm_backend}
• Unsupported lmdeploy device type: {device_type}

AI-assisted analysis of opendatalab/MinerU@4fe4bde114 (2026-08-14). Data as JSON: /api/errors/4c5b130e60201860. Report an issue: GitHub.