Merge branch 'litellm_cost_shard_harness_extensions' into litellm_cost_shard_responses_messages

This commit is contained in:
kerry 2026-09-19 19:53:49 +00:00
commit 22d3441789
3 changed files with 10 additions and 5 deletions

View file

@ -559,7 +559,7 @@
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[fireworks_ai-accounts-fireworks-models-deepseek-v4p1-flash-input_text]": [
"quota_management.spend_tracking.cost_matrix.logs_cost"
],
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[fireworks_ai-accounts-fireworks-models-deepseek-v4p1-flash-fallback_cache_read_at_input_rate]": [
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[fireworks_ai-accounts-fireworks-models-deepseek-v4p1-flash-fallback_cache_read_at_half_input_rate]": [
"quota_management.spend_tracking.cost_matrix.logs_cost"
],
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[fireworks_ai-accounts-fireworks-models-deepseek-v4p1-flash-stream]": [

View file

@ -296,7 +296,11 @@ def data_errors() -> tuple[str, ...]:
for case in CASES
if (
isinstance(case.expected, FailureExpected)
and (not isinstance(case.response, JsonResponse) or case.response.status < 400)
and (
not isinstance(case.response, JsonResponse)
or not 400 <= case.response.status <= 599
or not 400 <= case.expected.failure.status <= 599
)
)
or (
not isinstance(case.expected, FailureExpected)

View file

@ -7958,7 +7958,7 @@
}
},
{
"name": "fireworks_ai-accounts-fireworks-models-deepseek-v4p1-flash-fallback_cache_read_at_input_rate",
"name": "fireworks_ai-accounts-fireworks-models-deepseek-v4p1-flash-fallback_cache_read_at_half_input_rate",
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
"model": "fireworks_ai/accounts/fireworks/models/deepseek-v4p1-flash",
"request": {
@ -8014,9 +8014,10 @@
}
},
"expected": {
"spend": 0.0021672,
"input_cost": 0.0019392,
"spend": 0.0012456,
"input_cost": 0.0010176,
"output_cost": 0.000228,
"cache_read_cost": 0.0009216,
"prompt_tokens": 12928,
"completion_tokens": 380
}