d417ba2a48
Salvaged from PR #1563 by @kshitijk4poor. Cherry-picked with authorship preserved. - Route-aware pricing architecture replacing static MODEL_PRICING + heuristics - Canonical usage normalization (Anthropic/OpenAI/Codex API shapes) - Cache-aware billing (separate cache_read/cache_write rates) - Cost status tracking (estimated/included/unknown/actual) - OpenRouter live pricing via models API - Schema migration v4→v5 with billing metadata columns - Removed speculative forward-looking entries - Removed cost display from CLI status bar - Threaded OpenRouter metadata pre-warm Co-authored-by: kshitij <82637225+kshitijk4poor@users.noreply.github.com>
102 lines
3.0 KiB
Python
102 lines
3.0 KiB
Python
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"
|