huggingface/candle
1
0
Fork 0
mirror of https://github.com/huggingface/candle.git synced 2025-06-16 02:38:10 +00:00
Code Issues Packages Projects Releases Wiki Activity

Cuda acceleration for quantized model. (#1754)

Browse Source 
* Boilerplate for the quantized cuda support.

* More basic cuda support.

* More cuda quantization (quantize on cpu for now).

* Add the dequantization bit.

* Start adding some dedicated cuda kernels from llama.cpp.

* Move the kernel code.

* Start interfacing with the kernel.

* Tweak the kernel launch params.

* Bugfix for quantized metal.

* Fix some clippy lints.

* Tweak the launch parameters.

* Tweak cuda basics to perform a quantized matmul.

* Perform the dequantization on the cpu + use cublas for matmul.

* Add the dequantization kernel.

* Test the qmatmul.

* More kernels.

* Matmul-vec kernel.

* Add a couple kernels.

* More dequantization kernels.
This commit is contained in:
Laurent Mazare
2024-02-25 18:11:47 +01:00
committed by GitHub
parent 8d04f70f4d
commit 2f22afd80e
11 changed files with 1996 additions and 69 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
candle-kernels/src/lib.rs
View File

@ -4,6 +4,7 @@ pub const CAST: &str = include_str!(concat!(env!("OUT_DIR"), "/cast.ptx"));
pub const CONV: &str = include_str!(concat!(env!("OUT_DIR"), "/conv.ptx"));
pub const FILL: &str = include_str!(concat!(env!("OUT_DIR"), "/fill.ptx"));
pub const INDEXING: &str = include_str!(concat!(env!("OUT_DIR"), "/indexing.ptx"));
pub const QUANTIZED: &str = include_str!(concat!(env!("OUT_DIR"), "/quantized.ptx"));
pub const REDUCE: &str = include_str!(concat!(env!("OUT_DIR"), "/reduce.ptx"));
pub const TERNARY: &str = include_str!(concat!(env!("OUT_DIR"), "/ternary.ptx"));
pub const UNARY: &str = include_str!(concat!(env!("OUT_DIR"), "/unary.ptx"));

1536
candle-kernels/src/quantized.cu Normal file
View File

File diff suppressed because it is too large Load Diff