feat(npu): add OpenVINO reranker prototype

This commit is contained in:
William Valentin
2026-06-04 13:07:51 -07:00
parent 0a6f84fbf3
commit 0683253157
6 changed files with 1027 additions and 0 deletions
@@ -0,0 +1,19 @@
[Unit]
Description=OpenVINO NPU Reranker HTTP Service (port 18818)
After=network-online.target
[Service]
Type=simple
WorkingDirectory=/home/will/lab/swarm/openvino-reranker-npu
Environment=OPENVINO_RERANKER_HOST=127.0.0.1
Environment=OPENVINO_RERANKER_PORT=18818
Environment=OPENVINO_RERANKER_MODEL=cross-encoder/ms-marco-MiniLM-L6-v2
Environment=OPENVINO_RERANKER_MODEL_DIR=/home/will/.cache/openvino-models/rerankers/ms-marco-MiniLM-L6-v2-int8-ov
Environment=OPENVINO_RERANKER_DEVICE=NPU
Environment=OPENVINO_RERANKER_MAX_LENGTH=512
ExecStart=/home/will/.venvs/openvino-reranker/bin/python /home/will/lab/swarm/openvino-reranker-npu/server.py
Restart=on-failure
RestartSec=5
[Install]
WantedBy=default.target