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

[Model] PP support for Mamba-like models (#10992)

Browse Source 
Signed-off-by: mzusman <mor.zusmann@gmail.com>
This commit is contained in:
Mor Zusman
2024-12-11 04:53:37 +02:00
committed by GitHub
parent d5c5154fcf
commit ffa48c9146
11 changed files with 227 additions and 79 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
tests/distributed/test_pipeline_parallel.py
View File

@@ -156,13 +156,13 @@ TEXT_GENERATION_MODELS = {
# "internlm/internlm-chat-7b": PPTestSettings.fast(),
"internlm/internlm2-chat-7b": PPTestSettings.fast(trust_remote_code=True),
"inceptionai/jais-13b-chat": PPTestSettings.fast(),
# TODO: Implement PP
# "ai21labs/AI21-Jamba-1.5-Mini": PPTestSettings.fast(),
"ai21labs/Jamba-tiny-dev": PPTestSettings.fast(),
"meta-llama/Meta-Llama-3-8B": PPTestSettings.detailed(),
"openbmb/MiniCPM-2B-sft-bf16": PPTestSettings.fast(trust_remote_code=True),
"openbmb/MiniCPM3-4B": PPTestSettings.fast(trust_remote_code=True),
# Uses Llama
# "mistralai/Mistral-7B-Instruct-v0.1": PPTestSettings.fast(),
"state-spaces/mamba-130m-hf": PPTestSettings.fast(),
"mistralai/Mixtral-8x7B-Instruct-v0.1": PPTestSettings.fast(tp_base=4),
"mosaicml/mpt-7b": PPTestSettings.fast(),
"nvidia/Minitron-8B-Base": PPTestSettings.fast(),
@@ -234,6 +234,8 @@ TEST_MODELS = [
"OpenGVLab/InternVL2-1B",
"microsoft/Phi-3-vision-128k-instruct",
"fixie-ai/ultravox-v0_3",
# [LANGUAGE GENERATION - HYBRID ARCH]
"ai21labs/Jamba-tiny-dev",
]