hf_text-generation-inference/integration-tests/models/test_neox_sharded.py

47 lines
1.3 KiB
Python

import pytest
@pytest.fixture(scope="module")
def neox_sharded_handle(launcher):
with launcher(
"OpenAssistant/oasst-sft-1-pythia-12b", num_shard=2, use_flash_attention=False
) as handle:
yield handle
@pytest.fixture(scope="module")
async def neox_sharded(neox_sharded_handle):
await neox_sharded_handle.health()
return neox_sharded_handle.client
@pytest.mark.release
@pytest.mark.skip
@pytest.mark.asyncio
async def test_neox(neox_sharded, response_snapshot):
response = await neox_sharded.generate(
"<|prompter|>What is a meme, and what's the history behind this word?<|endoftext|><|assistant|>",
max_new_tokens=10,
decoder_input_details=True,
)
assert response.details.generated_tokens == 10
assert response == response_snapshot
@pytest.mark.release
@pytest.mark.skip
@pytest.mark.asyncio
async def test_neox_load(neox_sharded, generate_load, response_snapshot):
responses = await generate_load(
neox_sharded,
"<|prompter|>What is a meme, and what's the history behind this word?<|endoftext|><|assistant|>",
max_new_tokens=10,
n=4,
)
assert len(responses) == 4
assert all([r.generated_text == responses[0].generated_text for r in responses])
assert responses == response_snapshot