vrr/kvcache-ai-ktransformers
2
0
Fork 0
mirror of https://github.com/kvcache-ai/ktransformers.git synced 2025-09-10 06:14:58 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

[ADD] support multi-gpu qlen>1 q5_k

Browse source
This commit is contained in:
chenxl 2024-08-12 11:17:29 +00:00
parent f293803156
commit f5f79f5c0e
63 changed files with 3271 additions and 1285 deletions
Download patch file Download diff file Expand all files Collapse all files

6
ktransformers/ktransformers_ext/cuda/binding.cpp
View file

@ -3,8 +3,8 @@
* @Author : Azure-Tang
* @Date : 2024-07-25 13:38:30
* @Version : 1.0.0
* @LastEditors : Azure
* @LastEditTime : 2024-07-26 08:36:03
* @LastEditors : kkk1nak0
* @LastEditTime : 2024-08-09 01:45:02
* @Copyright (c) 2024 by KVCache.AI, All Rights Reserved.
**/
@ -23,6 +23,8 @@ PYBIND11_MODULE(KTransformersOps, m) {
py::arg("data"), py::arg("blk_size"), py::arg("device"));
m.def("dequantize_q6_k", &dequantize_q6_k, "Function to dequantize q6_k data.",
py::arg("data"), py::arg("blk_size"), py::arg("device"));
m.def("dequantize_q5_k", &dequantize_q5_k, "Function to dequantize q5_k data.",
py::arg("data"), py::arg("blk_size"), py::arg("device"));
m.def("dequantize_q4_k", &dequantize_q4_k, "Function to dequantize q4_k data.",
py::arg("data"), py::arg("blk_size"), py::arg("device"));
m.def("gptq_marlin_gemm", &gptq_marlin_gemm, "Function to perform GEMM using Marlin quantization.",