A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 14, 05:30 UTC
Use float32 for test_completion.py (#22385)
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Michael Goin committed
Aug 6, 2025 at 23:07 UTC
a00d8b236f515d8c29c6afc2ecb98aef22788ae1
1 file changed
+1
-2
tests/v1/entrypoints/openai/test_completion.py
+1
-2
@@ -20,9 +20,8 @@ MODEL_NAME = "facebook/opt-125m"
20
@pytest.fixture(scope="module")
21
def default_server_args():
22
return [
23
- # use half precision for speed and memory savings in CI environment
23
"--dtype",
25
- "bfloat16",
24
+ "float32",
25
"--max-model-len",
26
"2048",
27
"--max-num-seqs",