[Core] Pipeline Parallel Support (#4412)

Signed-off-by: Muralidhar Andoorveedu <muralidhar.andoorveedu@centml.ai>
This commit is contained in:
Murali Andoorveedu
2024-07-02 10:58:08 -07:00
committed by GitHub
parent 15aba081f3
commit c5832d2ae9
82 changed files with 1096 additions and 400 deletions
+2 -2
View File
@@ -14,7 +14,7 @@ import torch
from huggingface_hub import snapshot_download
from openai import BadRequestError
from ...utils import RemoteOpenAIServer
from ...utils import VLLM_PATH, RemoteOpenAIServer
# any model with a chat template should work here
MODEL_NAME = "HuggingFaceH4/zephyr-7b-beta"
@@ -77,7 +77,7 @@ def zephyr_lora_files():
@pytest.fixture(scope="module")
def ray_ctx():
ray.init()
ray.init(runtime_env={"working_dir": VLLM_PATH})
yield
ray.shutdown()
+2 -2
View File
@@ -16,7 +16,7 @@ from openai import BadRequestError
from vllm.transformers_utils.tokenizer import get_tokenizer
from ...utils import RemoteOpenAIServer
from ...utils import VLLM_PATH, RemoteOpenAIServer
# any model with a chat template should work here
MODEL_NAME = "HuggingFaceH4/zephyr-7b-beta"
@@ -79,7 +79,7 @@ def zephyr_lora_files():
@pytest.fixture(scope="module")
def ray_ctx():
ray.init()
ray.init(runtime_env={"working_dir": VLLM_PATH})
yield
ray.shutdown()
+2 -2
View File
@@ -5,14 +5,14 @@ import openai
import pytest
import ray
from ...utils import RemoteOpenAIServer
from ...utils import VLLM_PATH, RemoteOpenAIServer
EMBEDDING_MODEL_NAME = "intfloat/e5-mistral-7b-instruct"
@pytest.fixture(scope="module")
def ray_ctx():
ray.init()
ray.init(runtime_env={"working_dir": VLLM_PATH})
yield
ray.shutdown()
+2 -2
View File
@@ -6,7 +6,7 @@ import ray
# downloading lora to test lora requests
from huggingface_hub import snapshot_download
from ...utils import RemoteOpenAIServer
from ...utils import VLLM_PATH, RemoteOpenAIServer
# any model with a chat template should work here
MODEL_NAME = "HuggingFaceH4/zephyr-7b-beta"
@@ -22,7 +22,7 @@ def zephyr_lora_files():
@pytest.fixture(scope="module")
def ray_ctx():
ray.init()
ray.init(runtime_env={"working_dir": VLLM_PATH})
yield
ray.shutdown()
+2 -2
View File
@@ -24,13 +24,13 @@ TEST_IMAGE_URLS = [
@pytest.fixture(scope="module")
def ray_ctx():
ray.init()
ray.init(runtime_env={"working_dir": VLLM_PATH})
yield
ray.shutdown()
@pytest.fixture(scope="module")
def server():
def server(ray_ctx):
return RemoteOpenAIServer([
"--model",
MODEL_NAME,