#[non_exhaustive]pub enum GgmlType {
Show 33 variants
F32 = 0,
F16 = 1,
Q4_0 = 2,
Q4_1 = 3,
Q5_0 = 6,
Q5_1 = 7,
Q8_0 = 8,
Q8_1 = 9,
Q2K = 10,
Q3K = 11,
Q4K = 12,
Q5K = 13,
Q6K = 14,
Q8K = 15,
IQ2XXS = 16,
IQ2XS = 17,
IQ3XXS = 18,
IQ1S = 19,
IQ4NL = 20,
IQ3S = 21,
IQ2S = 22,
IQ4XS = 23,
I8 = 24,
I16 = 25,
I32 = 26,
I64 = 27,
F64 = 28,
IQ1M = 29,
BF16 = 30,
TQ1_0 = 34,
TQ2_0 = 35,
MXFP4 = 39,
NVFP4 = 40,
}Expand description
GGML tensor storage type (maps to ggml_type).
Used to set QuantizeParams::output_tensor_type and
QuantizeParams::token_embedding_type, and for per-tensor type overrides
in TensorTypeOverride.
Variants (Non-exhaustive)§
This enum is marked as non-exhaustive
Non-exhaustive enums could have additional variants added in future. Therefore, when matching against variants of non-exhaustive enums, an extra wildcard arm must be added to account for any future variants.
F32 = 0
F16 = 1
Q4_0 = 2
Q4_1 = 3
Q5_0 = 6
Q5_1 = 7
Q8_0 = 8
Q8_1 = 9
Q2K = 10
Q3K = 11
Q4K = 12
Q5K = 13
Q6K = 14
Q8K = 15
IQ2XXS = 16
IQ2XS = 17
IQ3XXS = 18
IQ1S = 19
IQ4NL = 20
IQ3S = 21
IQ2S = 22
IQ4XS = 23
I8 = 24
I16 = 25
I32 = 26
I64 = 27
F64 = 28
IQ1M = 29
BF16 = 30
TQ1_0 = 34
TQ2_0 = 35
MXFP4 = 39
NVFP4 = 40
Trait Implementations§
impl Copy for GgmlType
impl Eq for GgmlType
impl StructuralPartialEq for GgmlType
Auto Trait Implementations§
impl Freeze for GgmlType
impl RefUnwindSafe for GgmlType
impl Send for GgmlType
impl Sync for GgmlType
impl Unpin for GgmlType
impl UnsafeUnpin for GgmlType
impl UnwindSafe for GgmlType
Blanket Implementations§
Source§impl<T> BorrowMut<T> for Twhere
T: ?Sized,
impl<T> BorrowMut<T> for Twhere
T: ?Sized,
Source§fn borrow_mut(&mut self) -> &mut T
fn borrow_mut(&mut self) -> &mut T
Mutably borrows from an owned value. Read more