EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

[CI]Add EPLB CI. (#3568)

Browse Source 
### What this PR does / why we need it?
1.Add eplb ci to check the change of eplb feature.
2.Add param checking of eplb params. 
### Does this PR introduce _any_ user-facing change?

### How was this patch tested?
Qwen in A3.


- vLLM version: v0.11.0rc3
- vLLM main: https://github.com/vllm-project/vllm/commit/v0.11.0

---------

Signed-off-by: offline0806 <3337230449@qq.com>
Co-authored-by: offline0806 <3337230449@qq.com>
This commit is contained in:
offline893
2025-10-21 22:58:02 +08:00
committed by GitHub
parent 4c9af353ee
commit e916265b2b
11 changed files with 461 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
vllm_ascend/eplb/utils.py
View File

@@ -71,7 +71,7 @@ def model_register(model, model_config):
if config.model_type == "qwen3_moe":
model.num_moe_layers = config.num_hidden_layers
elif config.model_type == "deepseek_v2" or config.model_type == "deepseek_v3":
num_dense_layers = config.first_k_dense_replace
model.num_moe_layers = config.num_hidden_layers - num_dense_layers
model.num_dense_layers = config.first_k_dense_replace
model.num_moe_layers = config.num_hidden_layers - model.num_dense_layers
else:
raise NotImplementedError("EPLB is not supported.")