phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-17 16:42:44 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

take hugging face repo

Browse source
This commit is contained in:
Dinesh Yeduguru 2024-11-18 14:57:56 -08:00
parent acf9af841b
commit 8595b2af85
3 changed files with 25 additions and 16 deletions
Download patch file Download diff file Expand all files Collapse all files

14
llama_stack/providers/tests/inference/test_model_registration.py
View file

@ -6,8 +6,6 @@
import pytest
from llama_models.datatypes import CoreModelId
# How to run this test:
#
@ -28,10 +26,12 @@ class TestModelRegistration:
"remote::vllm",
"remote::tgi",
):
pytest.skip("70B instruct is too big only for local inference providers")
pytest.skip(
"Skipping test for remote inference providers since they can handle large models like 70B instruct"
)
# Try to register a model that's too large for local inference
with pytest.raises(Exception) as exc_info:
with pytest.raises(ValueError) as exc_info:
await models_impl.register_model(
model_id="Llama3.1-70B-Instruct",
)
@ -52,13 +52,13 @@ class TestModelRegistration:
_ = await models_impl.register_model(
model_id="custom-model",
metadata={"llama_model": CoreModelId.llama3_1_8b_instruct.value},
metadata={"llama_model": "meta-llama/Llama-2-7b"},
)
with pytest.raises(ValueError) as exc_info:
await models_impl.register_model(
model_id="custom-model-2",
metadata={"llama_model": CoreModelId.llama3_2_3b_instruct.value},
metadata={"llama_model": "meta-llama/Llama-2-7b"},
provider_model_id="custom-model",
)
@ -66,7 +66,7 @@ class TestModelRegistration:
async def test_register_with_invalid_llama_model(self, inference_stack):
_, models_impl = inference_stack
with pytest.raises(Exception) as exc_info:
with pytest.raises(ValueError) as exc_info:
await models_impl.register_model(
model_id="custom-model-2",
metadata={"llama_model": "invalid-llama-model"},