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, 22:15 UTC
[ci] Add AWS creds for AMD (#13572)
Kevin H. Luu committed
Feb 19, 2025 at 19:56 UTC
88f6ba3281f727d5641d362476ae68562b666081
2 files changed
+6
.buildkite/run-amd-test.sh
+4
@@ -121,6 +121,8 @@ if [[ $commands == *"--shard-id="* ]]; then
121
--rm \
122
-e HIP_VISIBLE_DEVICES="${GPU}" \
123
-e HF_TOKEN \
124
+ -e AWS_ACCESS_KEY_ID \
125
+ -e AWS_SECRET_ACCESS_KEY \
126
-v "${HF_CACHE}:${HF_MOUNT}" \
127
-e "HF_HOME=${HF_MOUNT}" \
128
--name "${container_name}_${GPU}" \
@@ -148,6 +150,8 @@ else
150
--rm \
151
-e HIP_VISIBLE_DEVICES=0 \
152
-e HF_TOKEN \
153
+ -e AWS_ACCESS_KEY_ID \
154
+ -e AWS_SECRET_ACCESS_KEY \
155
-v "${HF_CACHE}:${HF_MOUNT}" \
156
-e "HF_HOME=${HF_MOUNT}" \
157
--name "${container_name}" \
requirements-rocm.txt
+2
@@ -10,3 +10,5 @@ ray >= 2.10.0
10
peft
11
pytest-asyncio
12
tensorizer>=2.9.0
13
+runai-model-streamer==0.11.0
14
+runai-model-streamer-s3==0.11.0