sgl-project/sglang · error · ValueError

Inkling reasoning_effort must be finite and in [0.0, 0.99]

Error message

Inkling reasoning_effort must be finite and in [0.0, 0.99]

What it means

reasoning_effort must be a finite float within [0.0, 0.99]. NaN, infinity, negative values, or values >= 1.0 raise ValueError. The upper bound 0.99 leaves room for the renderer's own maximum-effort directive.

Source

Thrown at python/sglang/srt/parser/inkling_renderer.py:257

            if not isinstance(text, str):
                raise TypeError("Inkling thinking part payload must be a string")
            yield ("thinking", text)
        elif ptype in _IMAGE_PART_TYPES:
            yield ("image", "")
        elif ptype in _AUDIO_PART_TYPES:
            yield ("audio", "")
        else:
            raise ValueError(f"unsupported content part type: {ptype!r}")


def _format_reasoning_effort(reasoning_effort: float) -> str:
    if isinstance(reasoning_effort, bool) or not isinstance(
        reasoning_effort, (int, float)
    ):
        raise TypeError("Inkling reasoning_effort must be a number")
    value = float(reasoning_effort)
    if not math.isfinite(value) or not 0.0 <= value <= 0.99:
        raise ValueError("Inkling reasoning_effort must be finite and in [0.0, 0.99]")
    return f"{round(value, 2):g}"


def _expect_role(message: Mapping[str, Any]) -> str:
    role = message.get("role")
    if role not in ROLE_MESSAGE_TOKENS:
        raise ValueError(
            f"unsupported Inkling message role {role!r}; expected one of {sorted(ROLE_MESSAGE_TOKENS)}"
        )
    return str(role)


def _as_mapping(value: Any) -> Mapping[str, Any]:
    if isinstance(value, Mapping):
        return value
    if hasattr(value, "model_dump"):
        dumped = value.model_dump()
        if isinstance(dumped, Mapping):

View on GitHub (pinned to 0132848349)

Solutions

1. Clamp reasoning_effort to [0.0, 0.99], e.g. min(max(v,0.0),0.99)
2. Use 0.99 for maximum effort

Example fix

# before
reasoning_effort=1.0
# after
reasoning_effort=0.99

import math
if effort is not None:
    assert math.isfinite(effort) and 0.0 <= effort <= 0.99, effort

When it happens

Trigger: Passing reasoning_effort=1.0, 1.5, -0.1, float('nan'), or float('inf') to append_effort/chat kwargs.

Common situations: Users assuming a 0–1 or 0–100 scale; normalizing 'max' effort to 1.0.

Related errors

• Inkling reasoning_effort must be a number
• H.264 encoder did not return MP4 decoder config
• MXFP8 fused prologue requires head_dim-aligned Q/K/V.
• MXFP8 fused prologue requires K/V scale buffers.
• MXFP8 fused prologue requires interleaved K/V scale buffers

AI-assisted analysis of sgl-project/sglang@0132848349 (2026-08-28). Data as JSON: /api/errors/3ef740b2d8af4372. Report an issue: GitHub.