microsoft/semantic-kernel · error · NotSupportedException

The provided style '{style}' is not supported.

Error message

The provided style '{style}' is not supported.

What it means

The image style string is matched case-insensitively against only VIVID and NATURAL. Null is allowed (returns null). Any other value throws NotSupportedException before the request is sent. Style is a DALL-E 3-specific parameter.

Source

Thrown at dotnet/src/Connectors/Connectors.OpenAI/Core/ClientCore.TextToImage.cs:143

            "MEDIUM" => GeneratedImageQuality.MediumQuality,
            "LOW" => GeneratedImageQuality.LowQuality,
            "AUTO" => GeneratedImageQuality.Auto,
            _ => throw new NotSupportedException($"The provided quality '{quality}' is not supported.")
        };
    }

    private static GeneratedImageStyle? GetGeneratedImageStyle(string? style)
    {
        if (style is null)
        {
            return null;
        }

        return style.ToUpperInvariant() switch
        {
            "VIVID" => GeneratedImageStyle.Vivid,
            "NATURAL" => GeneratedImageStyle.Natural,
            _ => throw new NotSupportedException($"The provided style '{style}' is not supported.")
        };
    }

    private static GeneratedImageFormat? GetResponseFormat(object? responseFormat)
    {
        if (responseFormat is null)
        {
            return null;
        }

        if (responseFormat is GeneratedImageFormat format)
        {
            return format;
        }

        if (responseFormat is string formatString)
        {
            return formatString.ToUpperInvariant() switch

View on GitHub (pinned to c028a0c7dc)

Solutions

1. Use 'vivid' or 'natural' (case-insensitive), or leave Style unset for the API default.
2. If you need fine-grained stylistic control, express it in the text prompt instead of the style parameter.

Example fix

// before
var settings = new OpenAITextToImageExecutionSettings { Style = "cinematic" };

// after
var settings = new OpenAITextToImageExecutionSettings { Style = "vivid" };

static readonly HashSet<string> SupportedStyles = new(StringComparer.OrdinalIgnoreCase)
{
    "vivid", "natural"
};

void ValidateStyle(string? style)
{
    if (style is not null && !SupportedStyles.Contains(style.Trim()))
        throw new ArgumentOutOfRangeException(nameof(style),
            "Style must be 'vivid' or 'natural' (or null for default)");
}

try { await imageService.GetTextToImageAsync(prompt, settings); }
catch (NotSupportedException ex) when (ex.Message.Contains("style"))
{
    settings.Style = null;
    await imageService.GetTextToImageAsync(prompt, settings);
}

When it happens

Trigger: Passing an OpenAITextToImageExecutionSettings.Style value like 'artistic', 'photorealistic', or 'cinematic' that isn't in the two-value allow-list.

Common situations: Expecting a richer style taxonomy than the two DALL-E 3 options. Passing a style from a different image API (e.g. Stable Diffusion). Misspelling 'vivid' or 'natural'.

Related errors

• The provided quality '{quality}' is not supported.
• The provided response format '{formatString}' is not support
• The generated image has no valid content.
• Invalid image size: {size.width}x{size.height}.
• Prompt is required.

AI-assisted analysis of microsoft/semantic-kernel@c028a0c7dc (2026-08-13). Data as JSON: /api/errors/76edc07cda152367. Report an issue: GitHub.