vrr/kvcache-ai-ktransformers
2
0
Fork 0
mirror of https://github.com/kvcache-ai/ktransformers.git synced 2025-09-09 13:55:27 +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

2
ktransformers/tests/dequant_gpu_t.py
View file

@ -11,7 +11,7 @@ from ktransformers.operators.linear import KTransformerLinear, QuantizedLinearMa
from ktransformers.operators.experts import KTransformersMLPExpert, MLPExpertsTorch
from ktransformers.util.custom_gguf import GGUFLoader, dequantize_q4_k_gpu, dequantize_q4_k
import torch
import CudaOps
import KTransformersOps
torch.set_default_dtype(torch.bfloat16)
import time
from transformers import (