vllm-project/vllm · error · ValueError

Invalid format: '{value}'

Error message

Invalid format: '{value}'

What it means

_parse_size matches size strings against the regex ^\s*([\d.]+)\s*(gb|mb|kb|b)?\s*$ (case-insensitive). Anything else — "8 GB X", "0x1000", "8tb", "1,024GB", "half" — fails with ValueError('Invalid format'). Only gb/mb/kb/b units (binary multiples) plus bare byte counts are accepted.

Source

Thrown at vllm/distributed/kv_transfer/kv_connector/v1/mooncake/store/worker.py:213

    if not isinstance(value, str):
        try:
            return int(value)
        except (TypeError, ValueError) as e:
            raise TypeError(f"Unsupported type for size: {type(value)}") from e

    cleaned = value.strip().lower()
    if not cleaned:
        raise ValueError("Size cannot be empty.")

    unit_multipliers = {
        "gb": 1024**3,
        "mb": 1024**2,
        "kb": 1024,
        "b": 1,
    }
    match = re.match(r"^\s*([\d.]+)\s*(gb|mb|kb|b)?\s*$", cleaned)
    if not match:
        raise ValueError(f"Invalid format: '{value}'")

    number_str = match.group(1)
    unit = match.group(2) or "b"
    multiplier = unit_multipliers[unit]

    try:
        numeric_value = float(number_str)
    except ValueError as exc:
        raise ValueError(f"Invalid numeric value '{number_str}' in: '{value}'") from exc
    return int(numeric_value * multiplier)


def _align_up(value: int, alignment: int) -> int:
    return ((value + alignment - 1) // alignment) * alignment


def _estimate_disk_offload_staging_bytes(size_list: list[int]) -> int:
    data_size = sum(size_list)

View on GitHub (pinned to c794754062)

Solutions

1. Convert to a supported unit: "2048GB" instead of "2TB"
2. Remove commas/spaces: "1024GB" not "1,024 GB"
3. Or use a bare integer byte count

Example fix

// before
{ "global_segment_size": "2TB" }

// after
{ "global_segment_size": "2048GB" }

import re
SIZE_RE = re.compile(r"^[\d.]+\s*(gb|mb|kb|b)?$", re.I)
v = str(raw.get(key)).strip()
if v and not SIZE_RE.match(v):
    raise ValueError(f"{key} must look like '1024MB' or a bare byte count")

When it happens

Trigger: Using unsupported units (TB/PB), thousands separators, hex, or extra characters in global_segment_size/local_buffer_size.

Common situations: Users writing disk-style sizes ("2TB"), European decimal commas, or pasting values from cloud console quotas.

Related errors

• Unsupported type for size: {type(value)}
• Size cannot be empty.
• Invalid numeric value '{number_str}' in: '{value}'
• numa_bind_cpus entries must not be empty.
• Mooncake Transfer Engine initialization failed.

AI-assisted analysis of vllm-project/vllm@c794754062 (2026-08-14). Data as JSON: /api/errors/9c80f6ddf2b75b79. Report an issue: GitHub.