test: refactor tests and fix inference edge cases
- Convert protocol and MoE test classes to plain functions - Add real server/engine integration and generate_async tests - Isolate test model per test and use pytest tmp_path - Reset FastAPI engine state after inference tests - Fix generate_async StopIteration handling on Python 3.12 - Fix HF adapter MoE dense/shared and Gemma qk_norm mapping - Correct dev dependency httpx2 to httpx
This commit is contained in:
@@ -8,6 +8,13 @@ from fastapi.testclient import TestClient
|
||||
from astrai.inference import get_app
|
||||
|
||||
|
||||
@pytest.fixture(autouse=True)
|
||||
def _cleanup_app_engine():
|
||||
"""Reset the lazy FastAPI singleton engine after each inference test."""
|
||||
yield
|
||||
get_app().state.engine = None
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def client():
|
||||
"""Provide a test client for the FastAPI app."""
|
||||
|
||||
@@ -1,5 +1,6 @@
|
||||
"""Unit tests for GenerateResult accumulator and InferenceEngine.generate()."""
|
||||
|
||||
import asyncio
|
||||
import threading
|
||||
from unittest.mock import MagicMock, patch
|
||||
|
||||
@@ -156,6 +157,35 @@ def test_engine_generate_streaming_yields_tokens():
|
||||
assert tokens == ["t1", "t2"]
|
||||
|
||||
|
||||
def test_engine_generate_async_yields_tokens_until_stop():
|
||||
mock_model, mock_tokenizer = _make_engine_mocks(decode="tok")
|
||||
callbacks_saved = []
|
||||
|
||||
def capture_cb(prompt, **kw):
|
||||
callbacks_saved.append(kw.get("stream_callback"))
|
||||
|
||||
with patch("astrai.inference.engine.InferenceScheduler") as MockSched:
|
||||
instance = MockSched.return_value
|
||||
instance.add_task.side_effect = capture_cb
|
||||
instance.remove_task.return_value = []
|
||||
|
||||
eng = InferenceEngine(mock_model, mock_tokenizer, max_batch_size=1)
|
||||
agen = eng.generate_async("hello")
|
||||
|
||||
async def collect():
|
||||
out = []
|
||||
async for token in agen:
|
||||
out.append(token)
|
||||
return out
|
||||
|
||||
cb = callbacks_saved[0]
|
||||
cb("t1")
|
||||
cb("t2")
|
||||
cb(STOP)
|
||||
|
||||
assert asyncio.run(collect()) == ["t1", "t2"]
|
||||
|
||||
|
||||
def test_engine_generate_non_streaming_batch():
|
||||
mock_model, mock_tokenizer = _make_engine_mocks(decode="r")
|
||||
|
||||
|
||||
+227
-205
@@ -3,8 +3,6 @@
|
||||
import json
|
||||
from unittest.mock import MagicMock
|
||||
|
||||
import pytest
|
||||
|
||||
from astrai.inference.network.anthropic import AnthropicResponseBuilder
|
||||
from astrai.inference.network.openai import OpenAIResponseBuilder
|
||||
from astrai.inference.network.protocol import GenContext, StopChecker, StopInfo
|
||||
@@ -34,223 +32,247 @@ def _sse_payloads(events):
|
||||
return payloads
|
||||
|
||||
|
||||
class TestStopChecker:
|
||||
def test_check_finds_match(self):
|
||||
sc = StopChecker(["stop", "end"])
|
||||
assert sc.check("hello stop world") == "stop"
|
||||
|
||||
def test_check_returns_none_when_no_match(self):
|
||||
sc = StopChecker(["stop"])
|
||||
assert sc.check("hello world") is None
|
||||
|
||||
def test_check_empty_sequences(self):
|
||||
sc = StopChecker([])
|
||||
assert sc.check("hello") is None
|
||||
def _make_openai_builder():
|
||||
builder = OpenAIResponseBuilder()
|
||||
req = MagicMock()
|
||||
req.messages = [MagicMock(role="user", content="Hello")]
|
||||
req.stop = None
|
||||
req.model = "astrai"
|
||||
engine = MagicMock()
|
||||
engine.tokenizer.apply_chat_template.return_value = "Hello"
|
||||
builder.prepare(req, engine)
|
||||
return builder
|
||||
|
||||
|
||||
class TestGenContext:
|
||||
def test_defaults(self):
|
||||
ctx = GenContext(resp_id="a", created=1, model="m", prompt_tokens=10)
|
||||
assert ctx.completion_tokens == 0
|
||||
|
||||
def test_fields_mutable(self):
|
||||
ctx = GenContext(resp_id="a", created=1, model="m", prompt_tokens=10)
|
||||
ctx.completion_tokens = 42
|
||||
assert ctx.completion_tokens == 42
|
||||
def _make_anthropic_builder():
|
||||
builder = AnthropicResponseBuilder()
|
||||
req = MagicMock()
|
||||
req.messages = [MagicMock(role="user", content="Hello")]
|
||||
req.model = "claude"
|
||||
req.system = None
|
||||
engine = MagicMock()
|
||||
engine.tokenizer.apply_chat_template.return_value = "Hello"
|
||||
builder.prepare(req, engine)
|
||||
return builder
|
||||
|
||||
|
||||
class TestStopInfo:
|
||||
def test_defaults(self):
|
||||
s = StopInfo()
|
||||
assert s.matched is None
|
||||
assert s.body == ""
|
||||
assert s.yielded == ""
|
||||
|
||||
def test_with_values(self):
|
||||
s = StopInfo(matched="stop", body="hello stop", yielded="hello ")
|
||||
assert s.matched == "stop"
|
||||
assert s.body == "hello stop"
|
||||
assert s.yielded == "hello "
|
||||
def test_check_finds_match():
|
||||
sc = StopChecker(["stop", "end"])
|
||||
assert sc.check("hello stop world") == "stop"
|
||||
|
||||
|
||||
class TestOpenAIResponseBuilder:
|
||||
@pytest.fixture
|
||||
def builder(self):
|
||||
builder = OpenAIResponseBuilder()
|
||||
req = MagicMock()
|
||||
req.messages = [MagicMock(role="user", content="Hello")]
|
||||
req.stop = None
|
||||
req.model = "astrai"
|
||||
engine = MagicMock()
|
||||
engine.tokenizer.apply_chat_template.return_value = "Hello"
|
||||
builder.prepare(req, engine)
|
||||
return builder
|
||||
|
||||
def test_prepare_returns_prompt_ctx_stops(self, builder):
|
||||
req = MagicMock()
|
||||
req.messages = [MagicMock(role="user", content="Hi")]
|
||||
req.stop = ["END"]
|
||||
req.model = "gpt"
|
||||
engine = MagicMock()
|
||||
engine.tokenizer.apply_chat_template.return_value = "Hi"
|
||||
prompt, ctx, stops = builder.prepare(req, engine)
|
||||
assert prompt == "Hi"
|
||||
assert ctx.model == "gpt"
|
||||
assert ctx.prompt_tokens == 0
|
||||
assert stops == ["END"]
|
||||
|
||||
def test_prepare_no_stop_returns_empty_list(self, builder):
|
||||
req = MagicMock()
|
||||
req.messages = []
|
||||
req.stop = None
|
||||
req.model = "x"
|
||||
engine = MagicMock()
|
||||
engine.tokenizer.apply_chat_template.return_value = ""
|
||||
_, _, stops = builder.prepare(req, engine)
|
||||
assert stops == []
|
||||
|
||||
def test_format_stream_start(self, builder):
|
||||
ctx = _make_ctx()
|
||||
events = builder.format_stream_start(ctx)
|
||||
payloads = _sse_payloads(events)
|
||||
assert len(payloads) == 1
|
||||
p = payloads[0]
|
||||
assert p["object"] == "chat.completion.chunk"
|
||||
assert p["choices"][0]["delta"]["role"] == "assistant"
|
||||
assert p["choices"][0]["finish_reason"] is None
|
||||
|
||||
def test_format_chunk(self, builder):
|
||||
events = builder.format_chunk("hello", body="hello")
|
||||
payload = json.loads(events[0].split("data: ", 1)[1])
|
||||
assert payload["choices"][0]["delta"]["content"] == "hello"
|
||||
assert payload["choices"][0]["finish_reason"] is None
|
||||
|
||||
def test_format_stream_end(self, builder):
|
||||
ctx = _make_ctx(completion_tokens=5)
|
||||
stop = StopInfo(matched="stop")
|
||||
events = builder.format_stream_end(ctx, stop)
|
||||
payloads = _sse_payloads(events)
|
||||
finish = payloads[0]
|
||||
assert finish["choices"][0]["finish_reason"] == "stop"
|
||||
usage = payloads[1]
|
||||
assert usage["completion_tokens"] == 5
|
||||
assert usage["total_tokens"] == 15
|
||||
|
||||
def test_format_response(self, builder):
|
||||
ctx = _make_ctx()
|
||||
stop = StopInfo()
|
||||
resp = builder.format_response(ctx, "hello", stop)
|
||||
assert resp["object"] == "chat.completion"
|
||||
assert resp["choices"][0]["message"]["content"] == "hello"
|
||||
assert resp["usage"]["prompt_tokens"] == 10
|
||||
def test_check_returns_none_when_no_match():
|
||||
sc = StopChecker(["stop"])
|
||||
assert sc.check("hello world") is None
|
||||
|
||||
|
||||
class TestAnthropicResponseBuilder:
|
||||
@pytest.fixture
|
||||
def builder(self):
|
||||
builder = AnthropicResponseBuilder()
|
||||
req = MagicMock()
|
||||
req.messages = [MagicMock(role="user", content="Hello")]
|
||||
req.model = "claude"
|
||||
engine = MagicMock()
|
||||
engine.tokenizer.apply_chat_template.return_value = "Hello"
|
||||
req.system = None
|
||||
builder.prepare(req, engine)
|
||||
return builder
|
||||
def test_check_empty_sequences():
|
||||
sc = StopChecker([])
|
||||
assert sc.check("hello") is None
|
||||
|
||||
def test_prepare_messages(self, builder):
|
||||
req = MagicMock()
|
||||
req.messages = [MagicMock(role="user", content="Hi")]
|
||||
req.model = "claude"
|
||||
req.system = None
|
||||
req.stop_sequences = None
|
||||
engine = MagicMock()
|
||||
engine.tokenizer.apply_chat_template.return_value = "Hi"
|
||||
prompt, ctx, stops = builder.prepare(req, engine)
|
||||
assert prompt == "Hi"
|
||||
assert stops == []
|
||||
|
||||
def test_prepare_with_stop_sequences(self, builder):
|
||||
req = MagicMock()
|
||||
req.messages = []
|
||||
req.model = "x"
|
||||
req.stop_sequences = ["stop", "end"]
|
||||
req.system = None
|
||||
engine = MagicMock()
|
||||
engine.tokenizer.apply_chat_template.return_value = ""
|
||||
_, _, stops = builder.prepare(req, engine)
|
||||
assert stops == ["stop", "end"]
|
||||
def test_gen_context_defaults():
|
||||
ctx = GenContext(resp_id="a", created=1, model="m", prompt_tokens=10)
|
||||
assert ctx.completion_tokens == 0
|
||||
|
||||
def test_format_stream_start(self, builder):
|
||||
ctx = _make_ctx(prompt_tokens=3)
|
||||
events = builder.format_stream_start(ctx)
|
||||
payloads = _sse_payloads(events)
|
||||
assert len(payloads) == 2
|
||||
assert payloads[0]["type"] == "message_start"
|
||||
assert payloads[0]["message"]["usage"]["input_tokens"] == 3
|
||||
assert payloads[1]["type"] == "content_block_start"
|
||||
|
||||
def test_format_chunk(self, builder):
|
||||
events = builder.format_chunk("tok", body="tok")
|
||||
payload = json.loads(events[0].split("data: ", 1)[1])
|
||||
assert payload["type"] == "content_block_delta"
|
||||
assert payload["delta"]["text"] == "tok"
|
||||
def test_gen_context_fields_mutable():
|
||||
ctx = GenContext(resp_id="a", created=1, model="m", prompt_tokens=10)
|
||||
ctx.completion_tokens = 42
|
||||
assert ctx.completion_tokens == 42
|
||||
|
||||
def test_format_stream_end_no_stop(self, builder):
|
||||
ctx = _make_ctx(completion_tokens=3)
|
||||
stop = StopInfo()
|
||||
events = builder.format_stream_end(ctx, stop)
|
||||
payloads = _sse_payloads(events)
|
||||
# content_block_stop, message_delta, message_stop
|
||||
types = [p["type"] for p in payloads]
|
||||
assert types == ["content_block_stop", "message_delta", "message_stop"]
|
||||
assert payloads[1]["delta"]["stop_reason"] == "end_turn"
|
||||
|
||||
def test_format_stream_end_with_stop_trims_and_emits_remaining(self, builder):
|
||||
ctx = _make_ctx(completion_tokens=7)
|
||||
stop = StopInfo(
|
||||
matched="END",
|
||||
body="Hello world END extra",
|
||||
yielded="Hello ",
|
||||
)
|
||||
events = builder.format_stream_end(ctx, stop)
|
||||
payloads = _sse_payloads(events)
|
||||
# unyielded delta, content_block_stop, message_delta, message_stop
|
||||
types = [p["type"] for p in payloads]
|
||||
assert types == [
|
||||
"content_block_delta",
|
||||
"content_block_stop",
|
||||
"message_delta",
|
||||
"message_stop",
|
||||
]
|
||||
assert payloads[0]["delta"]["text"] == "world "
|
||||
assert payloads[2]["delta"]["stop_reason"] == "stop_sequence"
|
||||
assert payloads[2]["delta"]["stop_sequence"] == "END"
|
||||
def test_stop_info_defaults():
|
||||
s = StopInfo()
|
||||
assert s.matched is None
|
||||
assert s.body == ""
|
||||
assert s.yielded == ""
|
||||
|
||||
def test_format_stream_end_stop_trimmed_already_yielded(self, builder):
|
||||
ctx = _make_ctx()
|
||||
stop = StopInfo(
|
||||
matched="END",
|
||||
body="Hello END",
|
||||
yielded="Hello ",
|
||||
)
|
||||
events = builder.format_stream_end(ctx, stop)
|
||||
payloads = _sse_payloads(events)
|
||||
# No unyielded delta (everything already sent)
|
||||
types = [p["type"] for p in payloads]
|
||||
assert types == ["content_block_stop", "message_delta", "message_stop"]
|
||||
|
||||
def test_format_response_with_stop_trims_content(self, builder):
|
||||
ctx = _make_ctx()
|
||||
stop = StopInfo(matched="STOP", body="text STOP extra", yielded="text ")
|
||||
resp = builder.format_response(ctx, "text STOP extra", stop)
|
||||
assert resp["content"][0]["text"] == "text "
|
||||
assert resp["stop_reason"] == "stop_sequence"
|
||||
assert resp["stop_sequence"] == "STOP"
|
||||
def test_stop_info_with_values():
|
||||
s = StopInfo(matched="stop", body="hello stop", yielded="hello ")
|
||||
assert s.matched == "stop"
|
||||
assert s.body == "hello stop"
|
||||
assert s.yielded == "hello "
|
||||
|
||||
def test_format_response_no_stop(self, builder):
|
||||
ctx = _make_ctx()
|
||||
stop = StopInfo()
|
||||
resp = builder.format_response(ctx, "full text", stop)
|
||||
assert resp["content"][0]["text"] == "full text"
|
||||
assert resp["stop_reason"] == "end_turn"
|
||||
|
||||
def test_openai_prepare_returns_prompt_ctx_stops():
|
||||
builder = _make_openai_builder()
|
||||
req = MagicMock()
|
||||
req.messages = [MagicMock(role="user", content="Hi")]
|
||||
req.stop = ["END"]
|
||||
req.model = "gpt"
|
||||
engine = MagicMock()
|
||||
engine.tokenizer.apply_chat_template.return_value = "Hi"
|
||||
prompt, ctx, stops = builder.prepare(req, engine)
|
||||
assert prompt == "Hi"
|
||||
assert ctx.model == "gpt"
|
||||
assert ctx.prompt_tokens == 0
|
||||
assert stops == ["END"]
|
||||
|
||||
|
||||
def test_openai_prepare_no_stop_returns_empty_list():
|
||||
builder = _make_openai_builder()
|
||||
req = MagicMock()
|
||||
req.messages = []
|
||||
req.stop = None
|
||||
req.model = "x"
|
||||
engine = MagicMock()
|
||||
engine.tokenizer.apply_chat_template.return_value = ""
|
||||
_, _, stops = builder.prepare(req, engine)
|
||||
assert stops == []
|
||||
|
||||
|
||||
def test_openai_format_stream_start():
|
||||
builder = _make_openai_builder()
|
||||
ctx = _make_ctx()
|
||||
events = builder.format_stream_start(ctx)
|
||||
payloads = _sse_payloads(events)
|
||||
assert len(payloads) == 1
|
||||
p = payloads[0]
|
||||
assert p["object"] == "chat.completion.chunk"
|
||||
assert p["choices"][0]["delta"]["role"] == "assistant"
|
||||
assert p["choices"][0]["finish_reason"] is None
|
||||
|
||||
|
||||
def test_openai_format_chunk():
|
||||
builder = _make_openai_builder()
|
||||
events = builder.format_chunk("hello", body="hello")
|
||||
payload = json.loads(events[0].split("data: ", 1)[1])
|
||||
assert payload["choices"][0]["delta"]["content"] == "hello"
|
||||
assert payload["choices"][0]["finish_reason"] is None
|
||||
|
||||
|
||||
def test_openai_format_stream_end():
|
||||
builder = _make_openai_builder()
|
||||
ctx = _make_ctx(completion_tokens=5)
|
||||
stop = StopInfo(matched="stop")
|
||||
events = builder.format_stream_end(ctx, stop)
|
||||
payloads = _sse_payloads(events)
|
||||
finish = payloads[0]
|
||||
assert finish["choices"][0]["finish_reason"] == "stop"
|
||||
usage = payloads[1]
|
||||
assert usage["completion_tokens"] == 5
|
||||
assert usage["total_tokens"] == 15
|
||||
|
||||
|
||||
def test_openai_format_response():
|
||||
builder = _make_openai_builder()
|
||||
ctx = _make_ctx()
|
||||
stop = StopInfo()
|
||||
resp = builder.format_response(ctx, "hello", stop)
|
||||
assert resp["object"] == "chat.completion"
|
||||
assert resp["choices"][0]["message"]["content"] == "hello"
|
||||
assert resp["usage"]["prompt_tokens"] == 10
|
||||
|
||||
|
||||
def test_anthropic_prepare_messages():
|
||||
builder = _make_anthropic_builder()
|
||||
req = MagicMock()
|
||||
req.messages = [MagicMock(role="user", content="Hi")]
|
||||
req.model = "claude"
|
||||
req.system = None
|
||||
req.stop_sequences = None
|
||||
engine = MagicMock()
|
||||
engine.tokenizer.apply_chat_template.return_value = "Hi"
|
||||
prompt, ctx, stops = builder.prepare(req, engine)
|
||||
assert prompt == "Hi"
|
||||
assert stops == []
|
||||
|
||||
|
||||
def test_anthropic_prepare_with_stop_sequences():
|
||||
builder = _make_anthropic_builder()
|
||||
req = MagicMock()
|
||||
req.messages = []
|
||||
req.model = "x"
|
||||
req.stop_sequences = ["stop", "end"]
|
||||
req.system = None
|
||||
engine = MagicMock()
|
||||
engine.tokenizer.apply_chat_template.return_value = ""
|
||||
_, _, stops = builder.prepare(req, engine)
|
||||
assert stops == ["stop", "end"]
|
||||
|
||||
|
||||
def test_anthropic_format_stream_start():
|
||||
builder = _make_anthropic_builder()
|
||||
ctx = _make_ctx(prompt_tokens=3)
|
||||
events = builder.format_stream_start(ctx)
|
||||
payloads = _sse_payloads(events)
|
||||
assert len(payloads) == 2
|
||||
assert payloads[0]["type"] == "message_start"
|
||||
assert payloads[0]["message"]["usage"]["input_tokens"] == 3
|
||||
assert payloads[1]["type"] == "content_block_start"
|
||||
|
||||
|
||||
def test_anthropic_format_chunk():
|
||||
builder = _make_anthropic_builder()
|
||||
events = builder.format_chunk("tok", body="tok")
|
||||
payload = json.loads(events[0].split("data: ", 1)[1])
|
||||
assert payload["type"] == "content_block_delta"
|
||||
assert payload["delta"]["text"] == "tok"
|
||||
|
||||
|
||||
def test_anthropic_format_stream_end_no_stop():
|
||||
builder = _make_anthropic_builder()
|
||||
ctx = _make_ctx(completion_tokens=3)
|
||||
stop = StopInfo()
|
||||
events = builder.format_stream_end(ctx, stop)
|
||||
payloads = _sse_payloads(events)
|
||||
types = [p["type"] for p in payloads]
|
||||
assert types == ["content_block_stop", "message_delta", "message_stop"]
|
||||
assert payloads[1]["delta"]["stop_reason"] == "end_turn"
|
||||
|
||||
|
||||
def test_anthropic_format_stream_end_with_stop_trims_and_emits_remaining():
|
||||
builder = _make_anthropic_builder()
|
||||
ctx = _make_ctx(completion_tokens=7)
|
||||
stop = StopInfo(
|
||||
matched="END",
|
||||
body="Hello world END extra",
|
||||
yielded="Hello ",
|
||||
)
|
||||
events = builder.format_stream_end(ctx, stop)
|
||||
payloads = _sse_payloads(events)
|
||||
types = [p["type"] for p in payloads]
|
||||
assert types == [
|
||||
"content_block_delta",
|
||||
"content_block_stop",
|
||||
"message_delta",
|
||||
"message_stop",
|
||||
]
|
||||
assert payloads[0]["delta"]["text"] == "world "
|
||||
assert payloads[2]["delta"]["stop_reason"] == "stop_sequence"
|
||||
assert payloads[2]["delta"]["stop_sequence"] == "END"
|
||||
|
||||
|
||||
def test_anthropic_format_stream_end_stop_trimmed_already_yielded():
|
||||
builder = _make_anthropic_builder()
|
||||
ctx = _make_ctx()
|
||||
stop = StopInfo(
|
||||
matched="END",
|
||||
body="Hello END",
|
||||
yielded="Hello ",
|
||||
)
|
||||
events = builder.format_stream_end(ctx, stop)
|
||||
payloads = _sse_payloads(events)
|
||||
types = [p["type"] for p in payloads]
|
||||
assert types == ["content_block_stop", "message_delta", "message_stop"]
|
||||
|
||||
|
||||
def test_anthropic_format_response_with_stop_trims_content():
|
||||
builder = _make_anthropic_builder()
|
||||
ctx = _make_ctx()
|
||||
stop = StopInfo(matched="STOP", body="text STOP extra", yielded="text ")
|
||||
resp = builder.format_response(ctx, "text STOP extra", stop)
|
||||
assert resp["content"][0]["text"] == "text "
|
||||
assert resp["stop_reason"] == "stop_sequence"
|
||||
assert resp["stop_sequence"] == "STOP"
|
||||
|
||||
|
||||
def test_anthropic_format_response_no_stop():
|
||||
builder = _make_anthropic_builder()
|
||||
ctx = _make_ctx()
|
||||
stop = StopInfo()
|
||||
resp = builder.format_response(ctx, "full text", stop)
|
||||
assert resp["content"][0]["text"] == "full text"
|
||||
assert resp["stop_reason"] == "end_turn"
|
||||
|
||||
@@ -1,8 +1,15 @@
|
||||
"""Unit tests for the inference HTTP server."""
|
||||
|
||||
from pathlib import Path
|
||||
|
||||
import pytest
|
||||
import torch
|
||||
|
||||
from astrai.inference import get_app
|
||||
from astrai.inference.network.app import _create_engine
|
||||
from astrai.model.transformer import AutoRegressiveLM
|
||||
from astrai.serialization import save_model
|
||||
from tests.helpers import CHAT_TEMPLATE, build_test_tokenizer, make_tiny_config
|
||||
|
||||
|
||||
def test_health_no_model(client):
|
||||
@@ -212,5 +219,42 @@ def test_chat_completions_stop_sequence_stream(client, loaded_model):
|
||||
)
|
||||
|
||||
|
||||
def test_chat_completions_real_engine(tmp_path, client):
|
||||
"""POST /v1/chat/completions with a real tiny model and tokenizer."""
|
||||
cfg = make_tiny_config(vocab_size=256)
|
||||
model = AutoRegressiveLM(cfg).eval()
|
||||
save_model(cfg.to_dict(), model.state_dict(), str(tmp_path))
|
||||
|
||||
tokenizer = build_test_tokenizer(vocab_size=256, chat_template=CHAT_TEMPLATE)
|
||||
tokenizer.save_pretrained(str(tmp_path))
|
||||
|
||||
engine = _create_engine(
|
||||
Path(tmp_path),
|
||||
device="cpu",
|
||||
dtype=torch.float32,
|
||||
max_batch_size=1,
|
||||
max_seq_len=64,
|
||||
)
|
||||
try:
|
||||
get_app().state.engine = engine
|
||||
response = client.post(
|
||||
"/v1/chat/completions",
|
||||
json={
|
||||
"messages": [{"role": "user", "content": "Hello"}],
|
||||
"max_tokens": 4,
|
||||
"temperature": 0.0,
|
||||
"stream": False,
|
||||
},
|
||||
)
|
||||
assert response.status_code == 200
|
||||
data = response.json()
|
||||
content = data["choices"][0]["message"]["content"]
|
||||
assert isinstance(content, str)
|
||||
assert data["usage"]["completion_tokens"] > 0
|
||||
finally:
|
||||
engine.shutdown()
|
||||
get_app().state.engine = None
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
pytest.main([__file__, "-v"])
|
||||
|
||||
Reference in New Issue
Block a user