feat: Add Infinity rerank container with bge-reranker-base (CF-1317)
Local cross-encoder rerank service using BAAI/bge-reranker-base model via michaelf34/infinity:0.0.68. Serves /rerank endpoint on port 7998 for LiteLLM proxy integration. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
25
docker-compose.yml
Normal file
25
docker-compose.yml
Normal file
@@ -0,0 +1,25 @@
|
|||||||
|
services:
|
||||||
|
infinity-rerank:
|
||||||
|
image: michaelf34/infinity:0.0.68
|
||||||
|
container_name: infinity-rerank
|
||||||
|
command: v2 --model-id BAAI/bge-reranker-base --port 7998 --engine torch
|
||||||
|
ports:
|
||||||
|
- "7998:7998"
|
||||||
|
volumes:
|
||||||
|
- infinity-data:/app/.cache
|
||||||
|
environment:
|
||||||
|
- DO_NOT_TRACK=1
|
||||||
|
restart: unless-stopped
|
||||||
|
deploy:
|
||||||
|
resources:
|
||||||
|
limits:
|
||||||
|
memory: 4G
|
||||||
|
networks:
|
||||||
|
- agiliton-api
|
||||||
|
|
||||||
|
volumes:
|
||||||
|
infinity-data:
|
||||||
|
|
||||||
|
networks:
|
||||||
|
agiliton-api:
|
||||||
|
external: true
|
||||||
Reference in New Issue
Block a user