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, 13:30 UTC
Add hardware print to TPU V1 test (#16792)
Michael Goin committed
Apr 17, 2025 at 23:13 UTC
6a0f5475615e5dc1b21ef95f7c6d00d98a001a85
1 file changed
+3
-1
.buildkite/scripts/hardware_ci/run-tpu-v1-test.sh
+3
-1
@@ -17,10 +17,12 @@ source /etc/environment
17
docker run --privileged --net host --shm-size=16G -it \
18
-e "HF_TOKEN=$HF_TOKEN" --name tpu-test \
19
vllm-tpu /bin/bash -c "python3 -m pip install git+https://github.com/thuml/depyf.git \
20
- && python3 -m pip install pytest \
20
+ && python3 -m pip install pytest tpu-info \
21
&& python3 -m pip install lm_eval[api]==0.4.4 \
22
&& export VLLM_USE_V1=1 \
23
&& export VLLM_XLA_CHECK_RECOMPILATION=1 \
24
+ && echo HARDWARE \
25
+ && tpu-info \
26
&& echo TEST_0 \
27
&& pytest -v -s /workspace/vllm/tests/v1/tpu/test_perf.py \
28
&& echo TEST_1 \