kapsl-backends 0.1.1

Pluggable inference backends (ONNX, llama.cpp) for the Kapsl engine
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

pub mod engine_pool;
pub mod factory;
#[cfg(feature = "gguf-native")]
pub mod gguf_native;
#[cfg(feature = "native")]
pub mod native;
pub mod onnx;
pub mod onnx_classify;
pub mod onnx_embed;
#[cfg(feature = "onnx-cuda-pool")]
pub mod ort_pool_allocator;
mod provider_compat;

pub use engine_pool::{EnginePool, EnginePoolConfig};
pub use factory::{BackendFactory, OnnxRuntimeTuning};
#[cfg(feature = "gguf-native")]
pub use gguf_native::GgufNativeBackend;
#[cfg(feature = "native")]
pub use native::NativeBackend;
pub use onnx::OnnxBackend;
pub use onnx_classify::OnnxClassifyBackend;
pub use onnx_embed::OnnxEmbedBackend;