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

Simplify weight loading in Transformers backend (#21382)

Browse Source 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
Harry Mellor
2025-07-23 04:29:43 +01:00
committed by GitHub
parent 3ec7170ff1
commit f154bb9ff0
7 changed files with 53 additions and 76 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
tests/distributed/test_pipeline_parallel.py
View File

@@ -177,7 +177,7 @@ TEXT_GENERATION_MODELS = {
"ai21labs/Jamba-tiny-dev": PPTestSettings.fast(),
"meta-llama/Llama-3.2-1B-Instruct": PPTestSettings.detailed(),
# Tests TransformersForCausalLM
"ArthurZ/Ilama-3.2-1B": PPTestSettings.fast(),
"hmellor/Ilama-3.2-1B": PPTestSettings.fast(),
"openbmb/MiniCPM-2B-sft-bf16": PPTestSettings.fast(),
"openbmb/MiniCPM3-4B": PPTestSettings.fast(),
# Uses Llama
@@ -249,7 +249,7 @@ TEST_MODELS = [
# [LANGUAGE GENERATION]
"microsoft/Phi-3.5-MoE-instruct",
"meta-llama/Llama-3.2-1B-Instruct",
"ArthurZ/Ilama-3.2-1B",
"hmellor/Ilama-3.2-1B",
"ibm/PowerLM-3b",
"deepseek-ai/DeepSeek-V2-Lite-Chat",
# [LANGUAGE EMBEDDING]