mirror of
https://github.com/bentoml/OpenLLM.git
synced 2026-01-04 13:40:45 -05:00
7438005c04
* chore(config): simplify configuration and update start CLI output handling Signed-off-by: Aaron <29749331+aarnphm@users.noreply.github.com> * chore: remove state and message sent after server lifecycle Signed-off-by: Aaron <29749331+aarnphm@users.noreply.github.com> * chore: update color stream and refactor reusable logic Signed-off-by: Aaron <29749331+aarnphm@users.noreply.github.com> * chore: update documentations and mypy Signed-off-by: Aaron <29749331+aarnphm@users.noreply.github.com> --------- Signed-off-by: Aaron <29749331+aarnphm@users.noreply.github.com>
31 lines
902 B
Python
31 lines
902 B
Python
from __future__ import annotations
|
|
import os
|
|
import typing as t
|
|
|
|
import pytest
|
|
|
|
|
|
if t.TYPE_CHECKING:
|
|
import openllm
|
|
|
|
|
|
@pytest.mark.skipif(os.getenv('GITHUB_ACTIONS') is not None, reason='Model is too large for CI')
|
|
def test_flan_t5_implementation(prompt: str, llm: openllm.LLM[t.Any, t.Any]):
|
|
assert llm.generate(prompt)
|
|
|
|
assert llm.generate(prompt, temperature=0.8, top_p=0.23)
|
|
|
|
|
|
@pytest.mark.skipif(os.getenv('GITHUB_ACTIONS') is not None, reason='Model is too large for CI')
|
|
def test_opt_implementation(prompt: str, llm: openllm.LLM[t.Any, t.Any]):
|
|
assert llm.generate(prompt)
|
|
|
|
assert llm.generate(prompt, temperature=0.9, top_k=8)
|
|
|
|
|
|
@pytest.mark.skipif(os.getenv('GITHUB_ACTIONS') is not None, reason='Model is too large for CI')
|
|
def test_baichuan_implementation(prompt: str, llm: openllm.LLM[t.Any, t.Any]):
|
|
assert llm.generate(prompt)
|
|
|
|
assert llm.generate(prompt, temperature=0.95)
|