Files
memoryweave/deploy/systemd/bge-embed.service
xiaowei 2a350cf781 fix(distill): max_tokens 800→1200 — Agnes推理token预算不足
Agnes 2.0-flash是推理模型,推理token占大量预算。
800仍不够→JSON截断(parse error: unexpected end of JSON input)。
1200确保推理+正文都够用。
2026-08-20 01:13:06 +08:00

18 lines
625 B
Desktop File

[Unit]
Description=MemoryWeave BGE-M3 ONNX Embed Server
After=network.target
[Service]
Type=simple
Environment="BGE_MODEL_PATH=/home/muc/models/bge-m3/onnx"
Environment="BGE_PORT=8000"
ExecStartPre=/bin/bash -c 'if [ ! -f /home/muc/.hermes/scripts/bge_embed_server.py ]; then cp /tmp/memoryweave/deploy/bge_embed_server.py /home/muc/.hermes/scripts/ 2>/dev/null || true; fi'
ExecStart=/home/muc/.hermes/hermes-agent/.venv/bin/python3 /home/muc/.hermes/scripts/bge_embed_server.py
Restart=always
RestartSec=5
StandardOutput=append:/tmp/bge_embed.log
StandardError=append:/tmp/bge_embed.log
[Install]
WantedBy=default.target