vrr/kvcache-ai-ktransformers
2
0
Fork 0
mirror of https://github.com/kvcache-ai/ktransformers.git synced 2026-05-05 23:50:14 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

Merge pull request #911 from kvcache-ai/patch_v0.2.3post2

Browse source 
🔧 update multi-gpu-fp8-linear and multi-gpu marlin yaml
This commit is contained in:
ZiWei Yuan 2025-03-17 15:09:11 +08:00 • committed by GitHub
commit 8a8311cb04
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 3 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

4
ktransformers/optimize/optimize_rules/DeepSeek-V3-Chat-multi-gpu-fp8-linear-ggml-experts.yaml
View file

@ -10,7 +10,7 @@
name: "^model\\.layers\\.(0|[1-9]|[12][0-9])\\."
class: ktransformers.models.modeling_deepseek_v3.DeepseekV3RotaryEmbedding
replace:
class: ktransformers.operators.RoPE.KMoEGateDeepSeekV3
class: ktransformers.operators.RoPE.YarnRotaryEmbeddingV3
kwargs:
generate_device: "cuda:0"
prefill_device: "cuda:0"
@ -18,7 +18,7 @@
name: "^model\\.layers\\.([3456][0-9])\\."
class: ktransformers.models.modeling_deepseek_v3.DeepseekV3RotaryEmbedding
replace:
class: ktransformers.operators.RoPE.KMoEGateDeepSeekV3
class: ktransformers.operators.RoPE.YarnRotaryEmbeddingV3
kwargs:
generate_device: "cuda:1"
prefill_device: "cuda:1"

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

@ -10,7 +10,7 @@
name: "^model\\.layers\\.(0|[1-9]|[12][0-9])\\."
class: ktransformers.models.modeling_deepseek_v3.DeepseekV3RotaryEmbedding
replace:
class: ktransformers.operators.RoPE.KMoEGateDeepSeekV3
class: ktransformers.operators.RoPE.YarnRotaryEmbeddingV3
kwargs:
generate_device: "cuda:0"
prefill_device: "cuda:0"