#[non_exhaustive]pub struct FitParams {
pub margins: Option<Vec<usize>>,
pub n_ctx_min: u32,
}Expand description
Configuration for automatic GPU/CPU layer fitting.
Passed to crate::Client::builder (or crate::Client::from_gguf) so
llama.cpp can probe available device memory and pick the optimal number
of layers to offload to GPU automatically, instead of requiring a manual
n_gpu_layers value.
Marked #[non_exhaustive]; build via Default::default() and chain the
with_* setters.
Fields (Non-exhaustive)§
This struct is marked as non-exhaustive
Non-exhaustive structs could have additional fields added in future. Therefore, non-exhaustive structs cannot be constructed in external crates using the traditional
Struct { .. } syntax; cannot be matched against without a wildcard ..; and struct update syntax will not work.margins: Option<Vec<usize>>Memory margin per device in bytes. If None, defaults to 1 GiB per device.
n_ctx_min: u32Minimum context size to preserve during fitting (default: 4096).
Implementations§
Trait Implementations§
Auto Trait Implementations§
impl Freeze for FitParams
impl RefUnwindSafe for FitParams
impl Send for FitParams
impl Sync for FitParams
impl Unpin for FitParams
impl UnsafeUnpin for FitParams
impl UnwindSafe for FitParams
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