# distill.yaml
server:
  port: 8080
  host: 0.0.0.0
  read_timeout: 30s
  write_timeout: 30s

embedding:
  provider: openai
  model: text-embedding-3-small
  batch_size: 100

dedup:
  threshold: 0.15
  method: agglomerative
  linkage: average

compress:
  enabled: true
  target_reduction: 0.5
  preserve_structure: true

summarize:
  enabled: true
  max_tokens: 4000
  preserve_recent: 5

cache:
  enabled: true
  backend: memory  # or redis
  max_size: 100MB
  ttl: 1h
  redis_url: redis://localhost:6379

retriever:
  backend: pinecone  # or qdrant
  index: my-index
  top_k: 50

auth:
  api_keys:
    - ${DISTILL_API_KEY}

telemetry:
  metrics:
    enabled: true
    port: 9090
  tracing:
    enabled: false