mirror of
https://github.com/BerriAI/litellm.git
synced 2026-09-24 00:52:24 +00:00
chore: merge main into litellm_cost_shard_audio_images
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
This commit is contained in:
kerry
commit
0afb9bb587
4 changed files with 1021 additions and 2 deletions
|
|
@ -1384,6 +1384,72 @@
|
|||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[gpt-image-next-images-edit]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[azure-text-embeddings-4-large-deployment]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[bedrock-cohere-embeddings-v4]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[bedrock-cohere-rerank-v4]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[bedrock-embeddings-titan-v2]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[cohere-embeddings-v5]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[cohere-rerank-v4-one]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[cohere-rerank-v4-three]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[cohere-rerank-v4-total-tokens-fallback]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[fireworks-embeddings-v1]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[gemini-embeddings-002]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[omni-moderations-next-list]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[omni-moderations-next-single]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[text-completions-openai-basic]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[text-completions-openai-n-best]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[text-completions-openai-stream-usage]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[text-embeddings-3-large-dimensions]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[text-embeddings-4-small-batch]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[text-embeddings-4-small-single]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[text-embeddings-4-small-token-array]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[together-completions-v1]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[together-embeddings-v1]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[vertex-embeddings-text-006]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
"tests/integration/cost_calculation/test_cost_tracking.py::test_case_bills_expected_cost[gpt-5.6-responses_cache_read]": [
|
||||
"quota_management.spend_tracking.cost_matrix.logs_cost"
|
||||
],
|
||||
|
|
|
|||
|
|
@ -35,7 +35,10 @@ class CostMapEntry(BaseModel):
|
|||
max_output_tokens: int | None = None
|
||||
supports_function_calling: bool | None = None
|
||||
input_cost_per_token: float | None = None
|
||||
input_cost_per_query: float | None = None
|
||||
output_cost_per_token: float | None = None
|
||||
output_vector_size: int | None = None
|
||||
input_cost_per_token_batches: float | None = None
|
||||
cache_read_input_token_cost: float | None = None
|
||||
cache_creation_input_token_cost: float | None = None
|
||||
cache_creation_input_token_cost_above_1hr: float | None = None
|
||||
|
|
@ -211,7 +214,8 @@ class CostTrackingTestCase(BaseModel):
|
|||
if provider == "openai"
|
||||
and (
|
||||
self.endpoint == "/v1/responses"
|
||||
or self.rates.mode in {"chat", "audio_transcription", "audio_speech", "image_generation"}
|
||||
or self.rates.mode
|
||||
in {"chat", "embedding", "moderation", "audio_transcription", "audio_speech", "image_generation"}
|
||||
)
|
||||
else "openai/responses"
|
||||
if provider == "openai"
|
||||
|
|
@ -249,8 +253,11 @@ _PROVIDER_PREFIXES: Final[Mapping[str, str]] = MappingProxyType(
|
|||
"bedrock": "bedrock",
|
||||
"bedrock_converse": "bedrock/converse",
|
||||
"deepgram": "deepgram",
|
||||
"text-completion-openai": "text-completion-openai",
|
||||
"cohere": "cohere",
|
||||
"vertex_ai-language-models": "vertex_ai",
|
||||
"vertex_ai-image-models": "vertex_ai",
|
||||
"vertex_ai-embedding-models": "vertex_ai",
|
||||
"gemini": "",
|
||||
"together_ai": "",
|
||||
"fireworks_ai": "",
|
||||
|
|
@ -275,12 +282,17 @@ _LITELLM_PARAMS: Final[Mapping[str, Mapping[str, str]]] = MappingProxyType(
|
|||
}
|
||||
),
|
||||
"deepgram": MappingProxyType({}),
|
||||
"text-completion-openai": MappingProxyType({}),
|
||||
"cohere": MappingProxyType({}),
|
||||
"vertex_ai-language-models": MappingProxyType(
|
||||
{"vertex_project": "cc-scripted-project", "vertex_location": "us-central1"}
|
||||
),
|
||||
"vertex_ai-image-models": MappingProxyType(
|
||||
{"vertex_project": "cc-scripted-project", "vertex_location": "us-central1"}
|
||||
),
|
||||
"vertex_ai-embedding-models": MappingProxyType(
|
||||
{"vertex_project": "cc-scripted-project", "vertex_location": "us-central1"}
|
||||
),
|
||||
"gemini": MappingProxyType({}),
|
||||
"together_ai": MappingProxyType({}),
|
||||
"fireworks_ai": MappingProxyType({}),
|
||||
|
|
|
|||
|
|
@ -491,6 +491,98 @@
|
|||
"litellm_provider": "bedrock",
|
||||
"mode": "image_generation",
|
||||
"output_cost_per_image": 0.045
|
||||
},
|
||||
"text-embedding-4-small": {
|
||||
"input_cost_per_token": 1.01e-06,
|
||||
"output_cost_per_token": 0,
|
||||
"litellm_provider": "openai",
|
||||
"mode": "embedding"
|
||||
},
|
||||
"text-embedding-3-large-next": {
|
||||
"input_cost_per_token": 1.02e-06,
|
||||
"output_cost_per_token": 0,
|
||||
"litellm_provider": "openai",
|
||||
"mode": "embedding"
|
||||
},
|
||||
"azure/text-embedding-4-large": {
|
||||
"input_cost_per_token": 1.03e-06,
|
||||
"output_cost_per_token": 0,
|
||||
"litellm_provider": "azure",
|
||||
"mode": "embedding"
|
||||
},
|
||||
"embed-v5": {
|
||||
"input_cost_per_token": 1.04e-06,
|
||||
"output_cost_per_token": 0,
|
||||
"litellm_provider": "cohere",
|
||||
"mode": "embedding"
|
||||
},
|
||||
"amazon.titan-embed-text-v2:0": {
|
||||
"input_cost_per_token": 1.05e-06,
|
||||
"output_cost_per_token": 0,
|
||||
"litellm_provider": "bedrock",
|
||||
"mode": "embedding"
|
||||
},
|
||||
"cohere.embed-english-v4": {
|
||||
"input_cost_per_token": 1.06e-06,
|
||||
"output_cost_per_token": 0,
|
||||
"litellm_provider": "bedrock",
|
||||
"mode": "embedding"
|
||||
},
|
||||
"text-embedding-006": {
|
||||
"input_cost_per_token": 1.07e-06,
|
||||
"output_cost_per_token": 0,
|
||||
"litellm_provider": "vertex_ai-embedding-models",
|
||||
"mode": "embedding"
|
||||
},
|
||||
"gemini/gemini-embedding-002": {
|
||||
"input_cost_per_token": 1.08e-06,
|
||||
"output_cost_per_token": 0,
|
||||
"litellm_provider": "gemini",
|
||||
"mode": "embedding"
|
||||
},
|
||||
"together_ai/together-embed-v1": {
|
||||
"input_cost_per_token": 1.09e-06,
|
||||
"output_cost_per_token": 0,
|
||||
"litellm_provider": "together_ai",
|
||||
"mode": "embedding"
|
||||
},
|
||||
"fireworks_ai/fireworks-embed-v1": {
|
||||
"input_cost_per_token": 1.1e-06,
|
||||
"output_cost_per_token": 0,
|
||||
"litellm_provider": "fireworks_ai",
|
||||
"mode": "embedding"
|
||||
},
|
||||
"rerank-v4": {
|
||||
"input_cost_per_token": 1.11e-06,
|
||||
"output_cost_per_token": 0,
|
||||
"input_cost_per_query": 0.0021,
|
||||
"litellm_provider": "cohere",
|
||||
"mode": "rerank"
|
||||
},
|
||||
"cohere.rerank-v4:0": {
|
||||
"input_cost_per_token": 1.12e-06,
|
||||
"output_cost_per_token": 0,
|
||||
"input_cost_per_query": 0.0022,
|
||||
"litellm_provider": "bedrock",
|
||||
"mode": "rerank"
|
||||
},
|
||||
"gpt-3.5-turbo-instruct-next": {
|
||||
"input_cost_per_token": 1.14e-06,
|
||||
"output_cost_per_token": 2.14e-06,
|
||||
"litellm_provider": "text-completion-openai",
|
||||
"mode": "completion"
|
||||
},
|
||||
"omni-moderation-next": {
|
||||
"input_cost_per_token": null,
|
||||
"output_cost_per_token": 0,
|
||||
"litellm_provider": "openai",
|
||||
"mode": "moderation"
|
||||
},
|
||||
"together_ai/meta-llama/Llama-3.3-70B-Instruct-Turbo": {
|
||||
"input_cost_per_token": 1.16e-06,
|
||||
"output_cost_per_token": 2.16e-06,
|
||||
"litellm_provider": "together_ai",
|
||||
"mode": "completion"
|
||||
}
|
||||
},
|
||||
"cases": [
|
||||
|
|
@ -26375,6 +26467,851 @@
|
|||
"breakdown_persisted": false
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "text-embeddings-4-small-single",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "text-embedding-4-small",
|
||||
"endpoint": "/v1/embeddings",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"input": "one embedding"
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"object": "list",
|
||||
"data": [
|
||||
{
|
||||
"object": "embedding",
|
||||
"embedding": [
|
||||
0.1,
|
||||
0.2,
|
||||
0.3
|
||||
],
|
||||
"index": 0
|
||||
}
|
||||
],
|
||||
"model": "text-embedding-4-small",
|
||||
"usage": {
|
||||
"prompt_tokens": 7,
|
||||
"total_tokens": 7
|
||||
}
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 7.07e-06,
|
||||
"input_cost": 7.07e-06,
|
||||
"output_cost": 0.0,
|
||||
"prompt_tokens": 7,
|
||||
"completion_tokens": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "text-embeddings-4-small-batch",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "text-embedding-4-small",
|
||||
"endpoint": "/v1/embeddings",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"input": [
|
||||
"one",
|
||||
"two",
|
||||
"three"
|
||||
]
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"object": "list",
|
||||
"data": [
|
||||
{
|
||||
"object": "embedding",
|
||||
"embedding": [
|
||||
0.1,
|
||||
0.2,
|
||||
0.3
|
||||
],
|
||||
"index": 0
|
||||
},
|
||||
{
|
||||
"object": "embedding",
|
||||
"embedding": [
|
||||
0.1,
|
||||
0.2,
|
||||
0.3
|
||||
],
|
||||
"index": 1
|
||||
},
|
||||
{
|
||||
"object": "embedding",
|
||||
"embedding": [
|
||||
0.1,
|
||||
0.2,
|
||||
0.3
|
||||
],
|
||||
"index": 2
|
||||
}
|
||||
],
|
||||
"model": "text-embedding-4-small",
|
||||
"usage": {
|
||||
"prompt_tokens": 21,
|
||||
"total_tokens": 21
|
||||
}
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 2.1210000000000002e-05,
|
||||
"input_cost": 2.1210000000000002e-05,
|
||||
"output_cost": 0.0,
|
||||
"prompt_tokens": 21,
|
||||
"completion_tokens": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "text-embeddings-4-small-token-array",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "text-embedding-4-small",
|
||||
"endpoint": "/v1/embeddings",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"input": [
|
||||
1,
|
||||
2,
|
||||
3,
|
||||
4
|
||||
]
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"object": "list",
|
||||
"data": [
|
||||
{
|
||||
"object": "embedding",
|
||||
"embedding": [
|
||||
0.1,
|
||||
0.2,
|
||||
0.3
|
||||
],
|
||||
"index": 0
|
||||
}
|
||||
],
|
||||
"model": "text-embedding-4-small",
|
||||
"usage": {
|
||||
"prompt_tokens": 9,
|
||||
"total_tokens": 9
|
||||
}
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 9.090000000000001e-06,
|
||||
"input_cost": 9.090000000000001e-06,
|
||||
"output_cost": 0.0,
|
||||
"prompt_tokens": 9,
|
||||
"completion_tokens": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "text-embeddings-3-large-dimensions",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "text-embedding-3-large-next",
|
||||
"endpoint": "/v1/embeddings",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"input": "large embedding",
|
||||
"dimensions": 3
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"object": "list",
|
||||
"data": [
|
||||
{
|
||||
"object": "embedding",
|
||||
"embedding": [
|
||||
0.1,
|
||||
0.2,
|
||||
0.3
|
||||
],
|
||||
"index": 0
|
||||
}
|
||||
],
|
||||
"model": "text-embedding-3-large-next",
|
||||
"usage": {
|
||||
"prompt_tokens": 8,
|
||||
"total_tokens": 8
|
||||
}
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 8.16e-06,
|
||||
"input_cost": 8.16e-06,
|
||||
"output_cost": 0.0,
|
||||
"prompt_tokens": 8,
|
||||
"completion_tokens": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "azure-text-embeddings-4-large-deployment",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "azure/text-embedding-4-large",
|
||||
"endpoint": "/v1/embeddings",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"input": "azure embedding"
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"object": "list",
|
||||
"data": [
|
||||
{
|
||||
"object": "embedding",
|
||||
"embedding": [
|
||||
0.1,
|
||||
0.2,
|
||||
0.3
|
||||
],
|
||||
"index": 0
|
||||
}
|
||||
],
|
||||
"model": "azure/text-embedding-4-large",
|
||||
"usage": {
|
||||
"prompt_tokens": 8,
|
||||
"total_tokens": 8
|
||||
}
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 8.24e-06,
|
||||
"input_cost": 8.24e-06,
|
||||
"output_cost": 0.0,
|
||||
"prompt_tokens": 8,
|
||||
"completion_tokens": 0
|
||||
},
|
||||
"deployment": {
|
||||
"model": "azure/cc-pinned-embedding-deployment",
|
||||
"base_model": "azure/text-embedding-4-large"
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "cohere-embeddings-v5",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "embed-v5",
|
||||
"endpoint": "/v1/embeddings",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"input": "cohere embedding",
|
||||
"input_type": "search_query"
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"id": "emb-1",
|
||||
"embeddings": {
|
||||
"float": [
|
||||
[
|
||||
0.1,
|
||||
0.2,
|
||||
0.3
|
||||
]
|
||||
]
|
||||
},
|
||||
"meta": {
|
||||
"billed_units": {
|
||||
"input_tokens": 11
|
||||
}
|
||||
}
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 1.144e-05,
|
||||
"input_cost": 1.144e-05,
|
||||
"output_cost": 0.0,
|
||||
"prompt_tokens": 11,
|
||||
"completion_tokens": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "bedrock-embeddings-titan-v2",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "amazon.titan-embed-text-v2:0",
|
||||
"endpoint": "/v1/embeddings",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"input": "titan embedding"
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"embedding": [
|
||||
0.1,
|
||||
0.2,
|
||||
0.3
|
||||
],
|
||||
"inputTextTokenCount": 10
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 1.05e-05,
|
||||
"input_cost": 1.05e-05,
|
||||
"output_cost": 0.0,
|
||||
"prompt_tokens": 10,
|
||||
"completion_tokens": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "bedrock-cohere-embeddings-v4",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "cohere.embed-english-v4",
|
||||
"endpoint": "/v1/embeddings",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"input": "bedrock cohere embedding"
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"embeddings": [
|
||||
[
|
||||
0.1,
|
||||
0.2,
|
||||
0.3
|
||||
]
|
||||
],
|
||||
"id": "emb-bedrock-cohere-1",
|
||||
"response_type": "embeddings_floats",
|
||||
"texts": [
|
||||
"bedrock cohere embedding"
|
||||
]
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 5.3e-06,
|
||||
"input_cost": 5.3e-06,
|
||||
"output_cost": 0.0,
|
||||
"prompt_tokens": 5,
|
||||
"completion_tokens": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "vertex-embeddings-text-006",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "text-embedding-006",
|
||||
"endpoint": "/v1/embeddings",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"input": "vertex embedding"
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"predictions": [
|
||||
{
|
||||
"embeddings": {
|
||||
"values": [
|
||||
0.1,
|
||||
0.2,
|
||||
0.3
|
||||
],
|
||||
"statistics": {
|
||||
"token_count": 7,
|
||||
"truncated": false
|
||||
}
|
||||
}
|
||||
}
|
||||
]
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 7.4899999999999994e-06,
|
||||
"input_cost": 7.4899999999999994e-06,
|
||||
"output_cost": 0.0,
|
||||
"prompt_tokens": 7,
|
||||
"completion_tokens": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "gemini-embeddings-002",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "gemini/gemini-embedding-002",
|
||||
"endpoint": "/v1/embeddings",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"input": "gemini embedding"
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"embeddings": [
|
||||
{
|
||||
"values": [
|
||||
0.1,
|
||||
0.2,
|
||||
0.3
|
||||
]
|
||||
}
|
||||
],
|
||||
"usageMetadata": {
|
||||
"promptTokenCount": 7,
|
||||
"totalTokenCount": 7
|
||||
}
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 3.24e-06,
|
||||
"input_cost": 3.24e-06,
|
||||
"output_cost": 0.0,
|
||||
"prompt_tokens": 3,
|
||||
"completion_tokens": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "together-embeddings-v1",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "together_ai/together-embed-v1",
|
||||
"endpoint": "/v1/embeddings",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"input": "together embedding"
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"object": "list",
|
||||
"data": [
|
||||
{
|
||||
"object": "embedding",
|
||||
"embedding": [
|
||||
0.1,
|
||||
0.2,
|
||||
0.3
|
||||
],
|
||||
"index": 0
|
||||
}
|
||||
],
|
||||
"model": "together-embed-v1",
|
||||
"usage": {
|
||||
"prompt_tokens": 7,
|
||||
"total_tokens": 7
|
||||
}
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 7.63e-06,
|
||||
"input_cost": 7.63e-06,
|
||||
"output_cost": 0.0,
|
||||
"prompt_tokens": 7,
|
||||
"completion_tokens": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "fireworks-embeddings-v1",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "fireworks_ai/fireworks-embed-v1",
|
||||
"endpoint": "/v1/embeddings",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"input": "fireworks embedding"
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"object": "list",
|
||||
"data": [
|
||||
{
|
||||
"object": "embedding",
|
||||
"embedding": [
|
||||
0.1,
|
||||
0.2,
|
||||
0.3
|
||||
],
|
||||
"index": 0
|
||||
}
|
||||
],
|
||||
"model": "fireworks-embed-v1",
|
||||
"usage": {
|
||||
"prompt_tokens": 7,
|
||||
"total_tokens": 7
|
||||
}
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 7.7e-06,
|
||||
"input_cost": 7.7e-06,
|
||||
"output_cost": 0.0,
|
||||
"prompt_tokens": 7,
|
||||
"completion_tokens": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "cohere-rerank-v4-one",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "rerank-v4",
|
||||
"endpoint": "/v1/rerank",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"query": "rank this",
|
||||
"documents": [
|
||||
"a",
|
||||
"b"
|
||||
]
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"id": "rr-$REQUEST_ID",
|
||||
"results": [
|
||||
{
|
||||
"index": 0,
|
||||
"relevance_score": 0.9
|
||||
}
|
||||
],
|
||||
"meta": {
|
||||
"api_version": {
|
||||
"version": "2"
|
||||
},
|
||||
"billed_units": {
|
||||
"search_units": 1
|
||||
}
|
||||
}
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 0.0021,
|
||||
"input_cost": 0.0021,
|
||||
"output_cost": 0.0,
|
||||
"prompt_tokens": 0,
|
||||
"completion_tokens": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "cohere-rerank-v4-three",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "rerank-v4",
|
||||
"endpoint": "/v1/rerank",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"query": "rank this",
|
||||
"documents": [
|
||||
"a long document",
|
||||
"another long document",
|
||||
"third long document"
|
||||
]
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"id": "rr-three-$REQUEST_ID",
|
||||
"results": [
|
||||
{
|
||||
"index": 0,
|
||||
"relevance_score": 0.9
|
||||
}
|
||||
],
|
||||
"meta": {
|
||||
"api_version": {
|
||||
"version": "2"
|
||||
},
|
||||
"billed_units": {
|
||||
"search_units": 3
|
||||
}
|
||||
}
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 0.0063,
|
||||
"input_cost": 0.0063,
|
||||
"output_cost": 0.0,
|
||||
"prompt_tokens": 0,
|
||||
"completion_tokens": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "cohere-rerank-v4-total-tokens-fallback",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "rerank-v4",
|
||||
"endpoint": "/v1/rerank",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"query": "rank this",
|
||||
"documents": [
|
||||
"fallback a",
|
||||
"fallback b"
|
||||
]
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"id": "rr-fallback-$REQUEST_ID",
|
||||
"results": [
|
||||
{
|
||||
"index": 0,
|
||||
"relevance_score": 0.8
|
||||
}
|
||||
],
|
||||
"meta": {
|
||||
"billed_units": {
|
||||
"total_tokens": 99
|
||||
}
|
||||
}
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 0.0,
|
||||
"input_cost": 0.0,
|
||||
"output_cost": 0.0,
|
||||
"prompt_tokens": 0,
|
||||
"completion_tokens": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "bedrock-cohere-rerank-v4",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "cohere.rerank-v4:0",
|
||||
"endpoint": "/v1/rerank",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"query": "rank this",
|
||||
"documents": [
|
||||
"a",
|
||||
"b"
|
||||
]
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"results": [
|
||||
{
|
||||
"index": 0,
|
||||
"relevanceScore": 0.9
|
||||
}
|
||||
],
|
||||
"response_id": "rr-3",
|
||||
"token_count": 1
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 0.0022,
|
||||
"input_cost": 0.0022,
|
||||
"output_cost": 0.0,
|
||||
"prompt_tokens": 0,
|
||||
"completion_tokens": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "text-completions-openai-basic",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "gpt-3.5-turbo-instruct-next",
|
||||
"endpoint": "/v1/completions",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"prompt": "complete this"
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"id": "cmpl-basic-$REQUEST_ID",
|
||||
"object": "text_completion",
|
||||
"choices": [
|
||||
{
|
||||
"text": "done",
|
||||
"index": 0,
|
||||
"finish_reason": "stop"
|
||||
}
|
||||
],
|
||||
"usage": {
|
||||
"prompt_tokens": 9,
|
||||
"completion_tokens": 4,
|
||||
"total_tokens": 13
|
||||
}
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 1.882e-05,
|
||||
"input_cost": 1.026e-05,
|
||||
"output_cost": 8.56e-06,
|
||||
"prompt_tokens": 9,
|
||||
"completion_tokens": 4
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "text-completions-openai-stream-usage",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "gpt-3.5-turbo-instruct-next",
|
||||
"endpoint": "/v1/completions",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"prompt": "complete this",
|
||||
"stream": true,
|
||||
"stream_options": {
|
||||
"include_usage": true
|
||||
}
|
||||
},
|
||||
"response": {
|
||||
"content_type": "text/event-stream",
|
||||
"frames": [
|
||||
"data: {\"id\": \"cmpl-$REQUEST_ID\", \"object\": \"text_completion\", \"created\": 1789789000, \"model\": \"gpt-3.5-turbo-instruct-next\", \"choices\": [{\"text\": \"done\", \"index\": 0, \"finish_reason\": null}], \"usage\": null}",
|
||||
"data: {\"id\": \"cmpl-$REQUEST_ID\", \"object\": \"text_completion\", \"created\": 1789789000, \"model\": \"gpt-3.5-turbo-instruct-next\", \"choices\": [], \"usage\": {\"prompt_tokens\": 9, \"completion_tokens\": 4, \"total_tokens\": 13}}",
|
||||
"data: [DONE]"
|
||||
]
|
||||
},
|
||||
"expected": {
|
||||
"spend": 1.882e-05,
|
||||
"input_cost": 1.026e-05,
|
||||
"output_cost": 8.56e-06,
|
||||
"prompt_tokens": 9,
|
||||
"completion_tokens": 4
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "text-completions-openai-n-best",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "gpt-3.5-turbo-instruct-next",
|
||||
"endpoint": "/v1/completions",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"prompt": "complete this twice",
|
||||
"n": 2
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"id": "cmpl-n-best-$REQUEST_ID",
|
||||
"object": "text_completion",
|
||||
"choices": [
|
||||
{
|
||||
"text": "done",
|
||||
"index": 0,
|
||||
"finish_reason": "stop"
|
||||
},
|
||||
{
|
||||
"text": "also done",
|
||||
"index": 1,
|
||||
"finish_reason": "stop"
|
||||
}
|
||||
],
|
||||
"usage": {
|
||||
"prompt_tokens": 9,
|
||||
"completion_tokens": 8,
|
||||
"total_tokens": 17
|
||||
}
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 2.738e-05,
|
||||
"input_cost": 1.026e-05,
|
||||
"output_cost": 1.712e-05,
|
||||
"prompt_tokens": 9,
|
||||
"completion_tokens": 8
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "together-completions-v1",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "together_ai/meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
||||
"endpoint": "/v1/completions",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"prompt": "together complete"
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"id": "cmpl-together-$REQUEST_ID",
|
||||
"object": "text_completion",
|
||||
"choices": [
|
||||
{
|
||||
"text": "done",
|
||||
"index": 0,
|
||||
"finish_reason": "stop"
|
||||
}
|
||||
],
|
||||
"usage": {
|
||||
"prompt_tokens": 9,
|
||||
"completion_tokens": 4,
|
||||
"total_tokens": 13
|
||||
}
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 1.908e-05,
|
||||
"input_cost": 1.0439999999999998e-05,
|
||||
"output_cost": 8.64e-06,
|
||||
"prompt_tokens": 9,
|
||||
"completion_tokens": 4
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "omni-moderations-next-single",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "omni-moderation-next",
|
||||
"endpoint": "/v1/moderations",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"input": "safe text"
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"id": "modr-single-$REQUEST_ID",
|
||||
"model": "omni-moderation-next",
|
||||
"results": [
|
||||
{
|
||||
"flagged": false,
|
||||
"categories": {},
|
||||
"category_scores": {}
|
||||
}
|
||||
]
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 0.0,
|
||||
"input_cost": 0.0,
|
||||
"output_cost": 0.0,
|
||||
"prompt_tokens": 0,
|
||||
"completion_tokens": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "omni-moderations-next-list",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
"model": "omni-moderation-next",
|
||||
"endpoint": "/v1/moderations",
|
||||
"request": {
|
||||
"model": "$MODEL",
|
||||
"input": [
|
||||
"safe text",
|
||||
"more safe text"
|
||||
]
|
||||
},
|
||||
"response": {
|
||||
"content_type": "application/json",
|
||||
"body": {
|
||||
"id": "modr-list-$REQUEST_ID",
|
||||
"model": "omni-moderation-next",
|
||||
"results": [
|
||||
{
|
||||
"flagged": false,
|
||||
"categories": {},
|
||||
"category_scores": {}
|
||||
},
|
||||
{
|
||||
"flagged": false,
|
||||
"categories": {},
|
||||
"category_scores": {}
|
||||
}
|
||||
]
|
||||
}
|
||||
},
|
||||
"expected": {
|
||||
"spend": 0.0,
|
||||
"input_cost": 0.0,
|
||||
"output_cost": 0.0,
|
||||
"prompt_tokens": 0,
|
||||
"completion_tokens": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"name": "gpt-5.6-responses_cache_read",
|
||||
"covers": "quota_management.spend_tracking.cost_matrix.logs_cost",
|
||||
|
|
|
|||
|
|
@ -220,7 +220,11 @@ def test_case_bills_expected_cost(gateway: Gateway, case: CostTrackingTestCase)
|
|||
)
|
||||
elif case.response.content_type == "application/json":
|
||||
header: Final = cast(str | None, response.headers.get("x-litellm-response-cost"))
|
||||
assert header is not None and approx_equal(float(header), expected.spend), (
|
||||
assert (
|
||||
(header is None or approx_equal(float(header), 0.0))
|
||||
if expected.spend == 0
|
||||
else (header is not None and approx_equal(float(header), expected.spend))
|
||||
), (
|
||||
f"{case.name}: x-litellm-response-cost {header} != expected {expected.spend}"
|
||||
)
|
||||
assert row.spend is not None and approx_equal(row.spend, expected.spend), (
|
||||
|
|
|
|||
Loading…
Add table
Reference in a new issue