vrr/kvcache-ai-ktransformers
2
0
Fork 0
mirror of https://github.com/kvcache-ai/ktransformers.git synced 2025-09-16 18:19:42 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 2

Fix rocm example yaml

Browse source
This commit is contained in:
Azure-Tang 2025-03-15 22:27:02 -04:00
parent 4a31237346
commit 85c32fdd10
1 changed files with 2 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

4
ktransformers/optimize/optimize_rules/rocm/DeepSeek-V3-Chat.yaml
View file

@ -12,7 +12,7 @@
replace:
class: ktransformers.operators.linear.KTransformersLinear # optimized Kernel on quantized data types
kwargs:
generate_device: "cuda"
generate_device: "cpu"
prefill_device: "cuda"
generate_op: "KLinearCPUInfer"
prefill_op: "KLinearTorch"
@ -23,7 +23,7 @@
replace:
class: ktransformers.operators.linear.KTransformersLinear # optimized Kernel on quantized data types
kwargs:
generate_device: "cpu"
generate_device: "cuda"
prefill_device: "cuda"
generate_op: "KLinearQ8"
prefill_op: "KLinearTorch"