biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

Add pipeline parallel support to TransformersModel (#12832)

Browse Source 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Signed-off-by: Isotr0py <2037008807@qq.com>
Co-authored-by: Isotr0py <2037008807@qq.com>
This commit is contained in:
Harry Mellor
2025-03-25 02:41:45 +00:00
committed by GitHub
parent 911c8eb000
commit 97cfa65df7
4 changed files with 244 additions and 87 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
tests/distributed/test_pipeline_parallel.py
View File

@@ -175,6 +175,8 @@ TEXT_GENERATION_MODELS = {
"inceptionai/jais-13b-chat": PPTestSettings.fast(),
"ai21labs/Jamba-tiny-dev": PPTestSettings.fast(),
"meta-llama/Llama-3.2-1B-Instruct": PPTestSettings.detailed(),
# Tests TransformersModel
"ArthurZ/Ilama-3.2-1B": PPTestSettings.fast(),
"openbmb/MiniCPM-2B-sft-bf16": PPTestSettings.fast(),
"openbmb/MiniCPM3-4B": PPTestSettings.fast(),
# Uses Llama
@@ -243,6 +245,7 @@ TEST_MODELS = [
# [LANGUAGE GENERATION]
"microsoft/Phi-3.5-MoE-instruct",
"meta-llama/Llama-3.2-1B-Instruct",
# "ArthurZ/Ilama-3.2-1B", NOTE: Uncomment after #13905
"ibm/PowerLM-3b",
# [LANGUAGE EMBEDDING]
"intfloat/e5-mistral-7b-instruct",