G2: Embedding 本地化 — 配置模力方舟 API

- 确认 bge-m3 模型已下载 /home/muc/models/bge-m3/
- ONNX BGE 服务使用 12.7GB RAM,机器撑不住,已停用
- VLLM_ENDPOINT → https://ai.gitee.com/v1/embeddings
- 新增 MOLIFANG_API_KEY 环境变量
- deploy/zhiyid.service 同步到 /etc/systemd/system/
- 全链路验证通过: encode 389ms → ANN 40ms → rerank 921ms → MMR 0ms
- 本地 vLLM 部署标记为 G2.5(延后,需安装 vllm 依赖)

待办: G2.5 安装 vllm + 部署本地 BGE (port 8000) — 需要下载 torch 等依赖,约 30min
This commit is contained in:
xiaowei
2026-05-30 02:08:30 +08:00
parent 4541b0165a
commit 1d54231fca
+2 -1
View File
@@ -11,7 +11,8 @@ Environment=PORT=7821
Environment=STORAGE_BACKEND=lancedb
Environment=SQLITE_PATH=/var/lib/memoryweave/memoryweave.db
Environment=API_KEY=zhiyi-dev-key-2026
Environment=VLLM_ENDPOINT=http://localhost:8000/v1/embeddings
Environment=VLLM_ENDPOINT=https://ai.gitee.com/v1/embeddings
Environment=MOLIFANG_API_KEY=3TSVVXRFFECE4TISXHGE1VXDAXBIPAP6O1VPJK18
Environment=LLM_ENDPOINT=http://127.0.0.1:3000/v1/chat/completions
Environment=LLM_MODEL=qwen/qwen3.5-122b-a10b
Environment=LLM_API_KEY=sk-0ExNiLblJvIWBDpkS50fwOBw4MmqLyKdHJK5iQtlw9dOMWBP