Skip to content

GitLab

Unverified Commit e7cfe0f6 authored by Pavithra Vijayakrishnan's avatar Pavithra Vijayakrishnan Committed by GitHub
Browse files

test: Remove completions endpoint from vllm tests. (#1575) 


Signed-off-by: default avatarPavithra Vijayakrishnan <160681768+pvijayakrish@users.noreply.github.com>
parent 2528ea2c
Hide whitespace changes
Inline Side-by-side
Showing with 5 additions and 10 deletions
tests/serve/test_dynamo_serve.py
View file @ e7cfe0f6
......@@ -87,10 +87,9 @@ deployment_graphs = {
module="graphs.agg:Frontend",
config="configs/agg.yaml",
directory="/workspace/examples/llm",
endpoints=["v1/chat/completions", "v1/completions"],
endpoints=["v1/chat/completions"],
response_handlers=[
chat_completions_response_handler,
completions_response_handler,
],
marks=[pytest.mark.gpu_1, pytest.mark.vllm],
),
......@@ -115,10 +114,9 @@ deployment_graphs = {
module="graphs.disagg:Frontend",
config="configs/disagg.yaml",
directory="/workspace/examples/llm",
endpoints=["v1/chat/completions", "v1/completions"],
endpoints=["v1/chat/completions"],
response_handlers=[
chat_completions_response_handler,
completions_response_handler,
],
marks=[pytest.mark.gpu_2, pytest.mark.vllm],
),
......@@ -129,10 +127,9 @@ deployment_graphs = {
module="graphs.agg_router:Frontend",
config="configs/agg_router.yaml",
directory="/workspace/examples/llm",
endpoints=["v1/chat/completions", "v1/completions"],
endpoints=["v1/chat/completions"],
response_handlers=[
chat_completions_response_handler,
completions_response_handler,
],
marks=[pytest.mark.gpu_1, pytest.mark.vllm],
# FIXME: This is a hack to allow deployments to start before sending any requests.
......@@ -147,10 +144,9 @@ deployment_graphs = {
module="graphs.disagg_router:Frontend",
config="configs/disagg_router.yaml",
directory="/workspace/examples/llm",
endpoints=["v1/chat/completions", "v1/completions"],
endpoints=["v1/chat/completions"],
response_handlers=[
chat_completions_response_handler,
completions_response_handler,
],
marks=[pytest.mark.gpu_2, pytest.mark.vllm],
# FIXME: This is a hack to allow deployments to start before sending any requests.
......@@ -165,10 +161,9 @@ deployment_graphs = {
module="graphs.agg:Frontend",
config="configs/agg.yaml",
directory="/workspace/examples/multimodal",
endpoints=["v1/chat/completions", "v1/completions"],
endpoints=["v1/chat/completions"],
response_handlers=[
chat_completions_response_handler,
completions_response_handler,
],
marks=[pytest.mark.gpu_2, pytest.mark.vllm],
),
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment