kvcache-ai/ktransformers
1
0
Fork 0
mirror of https://github.com/kvcache-ai/ktransformers.git synced 2026-03-14 18:37:23 +00:00
Code Issues Packages Projects Releases Wiki Activity

[fix](kt-kernel): drop the weights held in Python for loading weights operation in C++ (#1695)

Browse Source
This commit is contained in:
SCDESPERTATE
2025-12-12 11:42:33 +08:00
committed by GitHub
parent 1e69563363
commit 008de19e16
1 changed files with 3 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
kt-kernel/python/utils/llamafile.py
View File

@@ -217,3 +217,6 @@ class LlamafileMoEWrapper(BaseMoEWrapper):
# Load weights
self.cpu_infer.submit(self.moe.load_weights_task(physical_to_logical_map_cpu.data_ptr()))
self.cpu_infer.sync()
# Drop original weights after loading
self.weights_to_keep = None