Enum QuantScheme 

#[repr(u8)]
pub enum QuantScheme {
    QInt8 = 1,
    QInt4 = 2,
    Binary = 3,
    FP16Passthrough = 4,
}

Quantization scheme for embedding vectors

Variants

QInt8 = 1

8-bit signed integer quantization

• Range: -128 to 127
• Size reduction: 4x (F32 → Int8)
• Accuracy: Very high

QInt4 = 2

4-bit packed quantization (future)

• Range: -8 to 7
• Size reduction: 8x (F32 → 4-bit)
• Accuracy: High

Binary = 3

1-bit binary quantization (future)

• Range: -1 or 1 (sign-based)
• Size reduction: 32x (F32 → 1-bit)
• Accuracy: Moderate

FP16Passthrough = 4

FP16 passthrough (future)

• Range: Half-precision float
• Size reduction: 2x (F32 → F16)
• Accuracy: Very high

Implementations

impl QuantScheme

pub fn bytes_per_value(self) -> usize

Returns the expected bytes per value for this quantization scheme

pub fn compression_ratio(self) -> f32

Returns the compression ratio compared to F32 (4 bytes)

Trait Implementations

impl Clone for QuantScheme

fn clone(&self) -> QuantScheme

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for QuantScheme

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for QuantScheme

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl PartialEq for QuantScheme

fn eq(&self, other: &QuantScheme) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Serialize for QuantScheme

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl Copy for QuantScheme

impl Eq for QuantScheme

impl StructuralPartialEq for QuantScheme