vrr/kvcache-ai-ktransformers
2
0
Fork 0
mirror of https://github.com/kvcache-ai/ktransformers.git synced 2026-04-29 12:19:50 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

⚡ ready to publish

Browse source
This commit is contained in:
liam 2025-02-10 12:29:23 +08:00
parent f892d22849
commit 83401dbb3b
6 changed files with 157 additions and 19 deletions
Download patch file Download diff file Expand all files Collapse all files

2
ktransformers/optimize/optimize_rules/DeepSeek-V3-Chat.yaml
View file

@ -1,7 +1,7 @@
- match:
class: ktransformers.models.modeling_deepseek_v3.DeepseekV3RotaryEmbedding
replace:
class: ktransformers.operators.RoPE.RotaryEmbeddingV3
class: ktransformers.operators.RoPE.YarnRotaryEmbeddingV3
kwargs:
generate_device: "cuda"
prefill_device: "cuda"