api-optimizer/config.example.yaml at main · Malli88/api-optimizer · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
# API Optimizer Configuration
# Copy this file to config.yaml and customize

# API Keys - Using OpenRouter as unified provider
# OpenRouter provides access to all models through a single API key
api_keys:
  # OpenRouter API key (recommended - unified access to all models)
  openrouter: "${OPENROUTER_API_KEY}"  # Get from https://openrouter.ai/keys

  # Direct provider keys (optional, if not using OpenRouter)
  # openai: "${OPENAI_API_KEY}"
  # anthropic: "${ANTHROPIC_API_KEY}"

# OpenRouter Configuration
openrouter:
  api_base: "https://openrouter.ai/api/v1"
  extra_headers:
    HTTP-Referer: "https://agent-zero.ai/"
    X-Title: "Agent Zero"

# Response Cache Settings
cache:
  enabled: true
  backend: "sqlite"  # sqlite or memory
  db_path: ".api_optimizer/cache.db"
  default_ttl: 3600  # seconds (1 hour)
  max_entries: 10000
  semantic_matching:
    enabled: true
    threshold: 0.92  # similarity threshold (0-1)
    embedding_model: "openai/text-embedding-3-small"  # via OpenRouter

# Model Router Settings
router:
  enabled: true
  default_model: "openai/gpt-4o-mini"  # Cheap default for simple tasks
  cost_optimization: true

  # Model preferences by complexity (using OpenRouter model names)
  complexity_routing:
    simple:
      preferred: "openai/gpt-4o-mini"      # Fast & cheap ($0.15/$0.60 per 1M)
      fallback: "anthropic/claude-3.5-haiku"
    medium:
      preferred: "openai/gpt-4o-mini"      # Still cost-effective
      fallback: "anthropic/claude-3.5-sonnet"
    complex:
      preferred: "anthropic/claude-opus-4.5"  # Best reasoning
      fallback: "openai/gpt-4o"
    critical:
      preferred: "anthropic/claude-opus-4.5"  # Maximum capability
      fallback: "openai/gpt-4o"

  # Cost limits
  max_cost_per_request: 1.00  # USD, null for unlimited

# Cost Monitor Settings
monitor:
  enabled: true
  db_path: ".api_optimizer/usage.db"

  # Budget alerts (USD)
  budgets:
    daily: 10.00
    weekly: 50.00
    monthly: 150.00

  # Hard caps - requests blocked when exceeded
  hard_caps:
    daily: 20.00
    weekly: 100.00
    monthly: 300.00

  # Alert callbacks (optional)
  alert_webhook: null  # URL to POST alerts to
  alert_email: null    # Email for alerts

# Prompt Optimizer Settings
optimizer:
  enabled: true

  # Optimization features
  remove_redundant_phrases: true
  normalize_whitespace: true
  compress_examples: false  # Summarize long examples

  # Token estimation
  warn_threshold: 4000  # Warn if prompt exceeds this
  max_tokens: 100000    # Error if prompt exceeds this

# Logging
logging:
  level: "INFO"  # DEBUG, INFO, WARNING, ERROR
  file: null     # Log file path, null for stdout only