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 15, 04:45 UTC
[Bugfix] Fix CI bitsandbytes failure (#19969)
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
Jee Jee Li committed
Jun 23, 2025 at 21:30 UTC
a6e6604d32ead92a20a615312b7a83668d0f9d7f
1 file changed
+1
tests/quantization/test_bitsandbytes.py
+1
@@ -159,6 +159,7 @@ def test_4bit_bnb_embedding_model(
159
with vllm_runner(model_name,
160
task="embed",
161
dtype=dtype,
162
+ gpu_memory_utilization=0.5,
163
quantization="bitsandbytes") as vllm_model:
164
vllm_outputs = vllm_model.encode(example_prompts)
165
check_embeddings_close(