File size: 3,790 Bytes
b5b9c2e | 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 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 | from types import SimpleNamespace
from agent.usage_pricing import (
CanonicalUsage,
estimate_usage_cost,
get_pricing_entry,
normalize_usage,
)
def test_normalize_usage_anthropic_keeps_cache_buckets_separate():
usage = SimpleNamespace(
input_tokens=1000,
output_tokens=500,
cache_read_input_tokens=2000,
cache_creation_input_tokens=400,
)
normalized = normalize_usage(usage, provider="anthropic", api_mode="anthropic_messages")
assert normalized.input_tokens == 1000
assert normalized.output_tokens == 500
assert normalized.cache_read_tokens == 2000
assert normalized.cache_write_tokens == 400
assert normalized.prompt_tokens == 3400
def test_normalize_usage_openai_subtracts_cached_prompt_tokens():
usage = SimpleNamespace(
prompt_tokens=3000,
completion_tokens=700,
prompt_tokens_details=SimpleNamespace(cached_tokens=1800),
)
normalized = normalize_usage(usage, provider="openai", api_mode="chat_completions")
assert normalized.input_tokens == 1200
assert normalized.cache_read_tokens == 1800
assert normalized.output_tokens == 700
def test_openrouter_models_api_pricing_is_converted_from_per_token_to_per_million(monkeypatch):
monkeypatch.setattr(
"agent.usage_pricing.fetch_model_metadata",
lambda: {
"anthropic/claude-opus-4.6": {
"pricing": {
"prompt": "0.000005",
"completion": "0.000025",
"input_cache_read": "0.0000005",
"input_cache_write": "0.00000625",
}
}
},
)
entry = get_pricing_entry(
"anthropic/claude-opus-4.6",
provider="openrouter",
base_url="https://openrouter.ai/api/v1",
)
assert float(entry.input_cost_per_million) == 5.0
assert float(entry.output_cost_per_million) == 25.0
assert float(entry.cache_read_cost_per_million) == 0.5
assert float(entry.cache_write_cost_per_million) == 6.25
def test_estimate_usage_cost_marks_subscription_routes_included():
result = estimate_usage_cost(
"gpt-5.3-codex",
CanonicalUsage(input_tokens=1000, output_tokens=500),
provider="openai-codex",
base_url="https://chatgpt.com/backend-api/codex",
)
assert result.status == "included"
assert float(result.amount_usd) == 0.0
def test_estimate_usage_cost_refuses_cache_pricing_without_official_cache_rate(monkeypatch):
monkeypatch.setattr(
"agent.usage_pricing.fetch_model_metadata",
lambda: {
"google/gemini-2.5-pro": {
"pricing": {
"prompt": "0.00000125",
"completion": "0.00001",
}
}
},
)
result = estimate_usage_cost(
"google/gemini-2.5-pro",
CanonicalUsage(input_tokens=1000, output_tokens=500, cache_read_tokens=100),
provider="openrouter",
base_url="https://openrouter.ai/api/v1",
)
assert result.status == "unknown"
def test_custom_endpoint_models_api_pricing_is_supported(monkeypatch):
monkeypatch.setattr(
"agent.usage_pricing.fetch_endpoint_model_metadata",
lambda base_url, api_key=None: {
"zai-org/GLM-5-TEE": {
"pricing": {
"prompt": "0.0000005",
"completion": "0.000002",
}
}
},
)
entry = get_pricing_entry(
"zai-org/GLM-5-TEE",
provider="custom",
base_url="https://llm.chutes.ai/v1",
api_key="test-key",
)
assert float(entry.input_cost_per_million) == 0.5
assert float(entry.output_cost_per_million) == 2.0
|