xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-10 06:55:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Rocm][CI] Fix DeekSeek V2-Lite Accuracy CI (#29135)

Browse Source 
Signed-off-by: charlifu <charlifu@amd.com>
This commit is contained in:
Charlie Fu 2025-11-21 19:11:02 -06:00 committed by GitHub
parent d5dbdbfcb2
commit 9a3101b2ba
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 21 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
.buildkite/scripts/scheduled_integration_test/deepseek_v2_lite_ep_eplb.sh
View File

@ -17,7 +17,17 @@ wait_for_server() {
}
MODEL="deepseek-ai/DeepSeek-V2-lite"
# Set BACKENDS based on platform
if command -v rocm-smi &> /dev/null || [[ -d /opt/rocm ]] || [[ -n "${ROCM_PATH:-}" ]]; then
# ROCm platform
BACKENDS=("allgather_reducescatter")
# Disable MOE padding for ROCm since it is causing eplb to fail
export VLLM_ROCM_MOE_PADDING=0
else
# Non-ROCm platform (CUDA/other)
BACKENDS=("deepep_high_throughput" "deepep_low_latency")
fi
cleanup() {
if [[ -n "${SERVER_PID:-}" ]] && kill -0 "${SERVER_PID}" 2>/dev/null; then

9
.buildkite/scripts/scheduled_integration_test/qwen30b_a3b_fp8_block_ep.sh
View File

@ -17,7 +17,16 @@ wait_for_server() {
}
MODEL="QWen/Qwen3-30B-A3B-FP8"
# Set BACKENDS based on platform
if command -v rocm-smi &> /dev/null || [[ -d /opt/rocm ]] || [[ -n "${ROCM_PATH:-}" ]]; then
# ROCm platform
BACKENDS=("allgather_reducescatter")
# Disable MOE padding for ROCm since it is causing eplb to fail
export VLLM_ROCM_MOE_PADDING=0
else
# Non-ROCm platform (CUDA/other)
BACKENDS=("deepep_high_throughput" "deepep_low_latency")
fi
cleanup() {
if [[ -n "${SERVER_PID:-}" ]] && kill -0 "${SERVER_PID}" 2>/dev/null; then