scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-09-14 23:21:35 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 102

Merge pull request #40581 from BerriAI/litellm_registry_audit_2026_09_10

Browse source 
fix(model_prices): registry audit 2026-09-10, absorb open pricing PRs
This commit is contained in:
Mateo Wang 2026-09-10 07:43:29 -07:00 • committed by GitHub
commit c1a83fc005
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
5 changed files with 154 additions and 49 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

3
litellm/llms/jina_ai/rerank/transformation.py
View file

@ -157,9 +157,6 @@ class JinaAIRerankConfig(BaseRerankConfig):
billed_units: RerankBilledUnits | None = None,
model_info: ModelInfo | None = None,
) -> tuple[float, float]:
"""
Jina AI reranker is priced at $0.000000018 per token.
"""
if (
model_info is None
or "input_cost_per_token" not in model_info

60
litellm/model_prices_and_context_window_backup.json
View file

@ -30295,7 +30295,7 @@
"supports_none_reasoning_effort": true,
"default_reasoning_effort": "none",
"supports_xhigh_reasoning_effort": false,
"supports_minimal_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"gpt-5.1-2025-11-13": {
"cache_read_input_token_cost": 1.25e-07,
@ -30340,7 +30340,7 @@
"supports_none_reasoning_effort": true,
"default_reasoning_effort": "none",
"supports_xhigh_reasoning_effort": false,
"supports_minimal_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"gpt-5.1-chat-latest": {
"cache_read_input_token_cost": 1.25e-07,
@ -30432,7 +30432,7 @@
"supports_none_reasoning_effort": true,
"default_reasoning_effort": "none",
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"gpt-5.2-2025-12-11": {
"cache_read_input_token_cost": 1.75e-07,
@ -30478,7 +30478,7 @@
"supports_none_reasoning_effort": true,
"default_reasoning_effort": "none",
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"gpt-5.2-chat-latest": {
"cache_read_input_token_cost": 1.75e-07,
@ -31474,7 +31474,7 @@
"supports_none_reasoning_effort": true,
"default_reasoning_effort": "none",
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"input_cost_per_token_above_272k_tokens_flex": 2.5e-06,
"output_cost_per_token_above_272k_tokens_flex": 1.125e-05,
"cache_read_input_token_cost_above_272k_tokens_flex": 2.5e-07
@ -31526,7 +31526,7 @@
"supports_none_reasoning_effort": true,
"default_reasoning_effort": "none",
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"input_cost_per_token_above_272k_tokens_flex": 2.5e-06,
"output_cost_per_token_above_272k_tokens_flex": 1.125e-05,
"cache_read_input_token_cost_above_272k_tokens_flex": 2.5e-07
@ -31578,7 +31578,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": false,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"input_cost_per_token_above_272k_tokens_flex": 3e-05,
"output_cost_per_token_above_272k_tokens_flex": 0.000135
},
@ -31629,7 +31629,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": false,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"input_cost_per_token_above_272k_tokens_flex": 3e-05,
"output_cost_per_token_above_272k_tokens_flex": 0.000135
},
@ -33709,13 +33709,14 @@
"supports_tool_choice": true
},
"jina-reranker-v2-base-multilingual": {
"input_cost_per_token": 1.8e-08,
"input_cost_per_token": 5e-08,
"litellm_provider": "jina_ai",
"max_input_tokens": 1024,
"max_output_tokens": 1024,
"max_tokens": 1024,
"mode": "rerank",
"output_cost_per_token": 1.8e-08
"output_cost_per_token": 0.0,
"source": "https://api.jina.ai/v1/models"
},
"jp.anthropic.claude-sonnet-4-5-20250929-v1:0": {
"cache_creation_input_token_cost": 4.125e-06,
@ -39969,6 +39970,45 @@
"supports_tool_choice": true,
"supports_vision": true
},
"openrouter/openai/gpt-5.6-sol": {
"cache_creation_input_token_cost": 2.5e-06,
"cache_creation_input_token_cost_above_272k_tokens": 5e-06,
"cache_read_input_token_cost": 2e-07,
"cache_read_input_token_cost_above_272k_tokens": 4e-07,
"default_reasoning_effort": "medium",
"input_cost_per_token": 2e-06,
"input_cost_per_token_above_272k_tokens": 4e-06,
"litellm_provider": "openrouter",
"max_input_tokens": 1050000,
"max_output_tokens": 128000,
"max_tokens": 128000,
"mode": "chat",
"output_cost_per_token": 1e-05,
"output_cost_per_token_above_272k_tokens": 1.5e-05,
"reasoning_effort_levels": [
"none",
"low",
"medium",
"high",
"xhigh",
"max"
],
"source": "https://openrouter.ai/openai/gpt-5.6-sol",
"supported_modalities": [
"text",
"image"
],
"supported_output_modalities": [
"text"
],
"supports_function_calling": true,
"supports_pdf_input": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": true,
"supports_tool_choice": true,
"supports_vision": true
},
"openrouter/openai/gpt-oss-120b": {
"input_cost_per_token": 3.7e-08,
"litellm_provider": "openrouter",

60
model_prices_and_context_window.json
View file

@ -30295,7 +30295,7 @@
"supports_none_reasoning_effort": true,
"default_reasoning_effort": "none",
"supports_xhigh_reasoning_effort": false,
"supports_minimal_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"gpt-5.1-2025-11-13": {
"cache_read_input_token_cost": 1.25e-07,
@ -30340,7 +30340,7 @@
"supports_none_reasoning_effort": true,
"default_reasoning_effort": "none",
"supports_xhigh_reasoning_effort": false,
"supports_minimal_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"gpt-5.1-chat-latest": {
"cache_read_input_token_cost": 1.25e-07,
@ -30432,7 +30432,7 @@
"supports_none_reasoning_effort": true,
"default_reasoning_effort": "none",
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"gpt-5.2-2025-12-11": {
"cache_read_input_token_cost": 1.75e-07,
@ -30478,7 +30478,7 @@
"supports_none_reasoning_effort": true,
"default_reasoning_effort": "none",
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"gpt-5.2-chat-latest": {
"cache_read_input_token_cost": 1.75e-07,
@ -31474,7 +31474,7 @@
"supports_none_reasoning_effort": true,
"default_reasoning_effort": "none",
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"input_cost_per_token_above_272k_tokens_flex": 2.5e-06,
"output_cost_per_token_above_272k_tokens_flex": 1.125e-05,
"cache_read_input_token_cost_above_272k_tokens_flex": 2.5e-07
@ -31526,7 +31526,7 @@
"supports_none_reasoning_effort": true,
"default_reasoning_effort": "none",
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"input_cost_per_token_above_272k_tokens_flex": 2.5e-06,
"output_cost_per_token_above_272k_tokens_flex": 1.125e-05,
"cache_read_input_token_cost_above_272k_tokens_flex": 2.5e-07
@ -31578,7 +31578,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": false,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"input_cost_per_token_above_272k_tokens_flex": 3e-05,
"output_cost_per_token_above_272k_tokens_flex": 0.000135
},
@ -31629,7 +31629,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": false,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"input_cost_per_token_above_272k_tokens_flex": 3e-05,
"output_cost_per_token_above_272k_tokens_flex": 0.000135
},
@ -33709,13 +33709,14 @@
"supports_tool_choice": true
},
"jina-reranker-v2-base-multilingual": {
"input_cost_per_token": 1.8e-08,
"input_cost_per_token": 5e-08,
"litellm_provider": "jina_ai",
"max_input_tokens": 1024,
"max_output_tokens": 1024,
"max_tokens": 1024,
"mode": "rerank",
"output_cost_per_token": 1.8e-08
"output_cost_per_token": 0.0,
"source": "https://api.jina.ai/v1/models"
},
"jp.anthropic.claude-sonnet-4-5-20250929-v1:0": {
"cache_creation_input_token_cost": 4.125e-06,
@ -39969,6 +39970,45 @@
"supports_tool_choice": true,
"supports_vision": true
},
"openrouter/openai/gpt-5.6-sol": {
"cache_creation_input_token_cost": 2.5e-06,
"cache_creation_input_token_cost_above_272k_tokens": 5e-06,
"cache_read_input_token_cost": 2e-07,
"cache_read_input_token_cost_above_272k_tokens": 4e-07,
"default_reasoning_effort": "medium",
"input_cost_per_token": 2e-06,
"input_cost_per_token_above_272k_tokens": 4e-06,
"litellm_provider": "openrouter",
"max_input_tokens": 1050000,
"max_output_tokens": 128000,
"max_tokens": 128000,
"mode": "chat",
"output_cost_per_token": 1e-05,
"output_cost_per_token_above_272k_tokens": 1.5e-05,
"reasoning_effort_levels": [
"none",
"low",
"medium",
"high",
"xhigh",
"max"
],
"source": "https://openrouter.ai/openai/gpt-5.6-sol",
"supported_modalities": [
"text",
"image"
],
"supported_output_modalities": [
"text"
],
"supports_function_calling": true,
"supports_pdf_input": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": true,
"supports_tool_choice": true,
"supports_vision": true
},
"openrouter/openai/gpt-oss-120b": {
"input_cost_per_token": 3.7e-08,
"litellm_provider": "openrouter",

63
tests/test_litellm/llms/openai/test_gpt5_transformation.py
View file

@ -276,6 +276,17 @@ def test_gpt5_drops_reasoning_effort_xhigh_when_requested(config: OpenAIConfig):
assert "reasoning_effort" not in params
def test_gpt5_1_gpt5_2_gpt5_4_drop_minimal_reasoning_effort(config: OpenAIConfig):
for model in ("gpt-5.1", "gpt-5.2", "gpt-5.4", "gpt-5.4-pro"):
params = config.map_openai_params(
non_default_params={"reasoning_effort": "minimal"},
optional_params={},
model=model,
drop_params=True,
)
assert "reasoning_effort" not in params
# GPT-5.1 temperature handling tests
def test_gpt5_1_model_detection(gpt5_config: OpenAIGPT5Config):
"""Test that models supporting reasoning_effort='none' are correctly detected via model map."""
@ -388,26 +399,26 @@ def test_gpt5_4_mini_allows_reasoning_effort_none(config: OpenAIConfig):
assert params["reasoning_effort"] == "none"
def test_gpt5_4_allows_reasoning_effort_minimal(config: OpenAIConfig):
"""gpt-5.4 supports reasoning_effort='minimal'."""
params = config.map_openai_params(
non_default_params={"reasoning_effort": "minimal"},
optional_params={},
model="gpt-5.4",
drop_params=False,
)
assert params["reasoning_effort"] == "minimal"
def test_gpt5_4_rejects_reasoning_effort_minimal(config: OpenAIConfig):
"""gpt-5.4 rejects reasoning_effort='minimal'."""
with pytest.raises(litellm.utils.UnsupportedParamsError):
config.map_openai_params(
non_default_params={"reasoning_effort": "minimal"},
optional_params={},
model="gpt-5.4",
drop_params=False,
)
def test_gpt5_4_pro_allows_reasoning_effort_minimal(config: OpenAIConfig):
"""gpt-5.4-pro supports reasoning_effort='minimal'."""
params = config.map_openai_params(
non_default_params={"reasoning_effort": "minimal"},
optional_params={},
model="gpt-5.4-pro",
drop_params=False,
)
assert params["reasoning_effort"] == "minimal"
def test_gpt5_4_pro_rejects_reasoning_effort_minimal(config: OpenAIConfig):
"""gpt-5.4-pro rejects reasoning_effort='minimal'."""
with pytest.raises(litellm.utils.UnsupportedParamsError):
config.map_openai_params(
non_default_params={"reasoning_effort": "minimal"},
optional_params={},
model="gpt-5.4-pro",
drop_params=False,
)
def test_gpt5_4_mini_rejects_reasoning_effort_minimal(config: OpenAIConfig):
@ -468,13 +479,13 @@ def test_gpt5_minimal_dict_triggers_validation(config: OpenAIConfig):
def test_gpt5_minimal_dict_accepted_for_supported_model(config: OpenAIConfig):
"""Dict with effort='minimal' passes through for gpt-5.4+."""
"""Dict with effort='minimal' passes through for gpt-5."""
params = config.map_openai_params(
non_default_params={
"reasoning_effort": {"effort": "minimal", "summary": "detailed"}
},
optional_params={},
model="gpt-5.4",
model="gpt-5",
drop_params=False,
)
assert params["reasoning_effort"] == "minimal"
@ -482,8 +493,8 @@ def test_gpt5_minimal_dict_accepted_for_supported_model(config: OpenAIConfig):
def test_gpt5_supports_reasoning_effort_level_minimal(gpt5_config: OpenAIGPT5Config):
"""Test that _supports_reasoning_effort_level correctly identifies minimal support."""
assert gpt5_config._supports_reasoning_effort_level("gpt-5.4", "minimal")
assert gpt5_config._supports_reasoning_effort_level("gpt-5.4-pro", "minimal")
assert not gpt5_config._supports_reasoning_effort_level("gpt-5.4", "minimal")
assert not gpt5_config._supports_reasoning_effort_level("gpt-5.4-pro", "minimal")
assert not gpt5_config._supports_reasoning_effort_level("gpt-5.4-mini", "minimal")
assert not gpt5_config._supports_reasoning_effort_level("gpt-5.4-nano", "minimal")
@ -504,10 +515,10 @@ def test_gpt5_minimal_explicitly_disabled_check(gpt5_config: OpenAIGPT5Config):
assert gpt5_config._is_reasoning_effort_level_explicitly_disabled(
"openai/gpt-5.4-mini", "minimal"
)
assert not gpt5_config._is_reasoning_effort_level_explicitly_disabled(
assert gpt5_config._is_reasoning_effort_level_explicitly_disabled(
"gpt-5.4", "minimal"
)
assert not gpt5_config._is_reasoning_effort_level_explicitly_disabled(
assert gpt5_config._is_reasoning_effort_level_explicitly_disabled(
"gpt-5.4-pro", "minimal"
)
@ -523,8 +534,8 @@ def test_is_explicitly_disabled_factory_minimal():
assert _is_explicitly_disabled_factory("gpt-5.4-mini", None, key)
assert _is_explicitly_disabled_factory("gpt-5.4-nano", None, key)
assert _is_explicitly_disabled_factory("openai/gpt-5.4-mini", None, key)
assert not _is_explicitly_disabled_factory("gpt-5.4", None, key)
assert not _is_explicitly_disabled_factory("gpt-5.4-pro", None, key)
assert _is_explicitly_disabled_factory("gpt-5.4", None, key)
assert _is_explicitly_disabled_factory("gpt-5.4-pro", None, key)
assert not _is_explicitly_disabled_factory("gpt-5.4-turbo-preview", None, key)

17
tests/test_litellm/test_cost_calculator.py
View file

@ -18,6 +18,7 @@ from litellm.cost_calculator import (
response_cost_calculator,
)
from litellm.types.llms.openai import OpenAIRealtimeStreamList
from litellm.types.rerank import RerankResponse
from litellm.types.utils import (
CacheCreationTokenDetails,
ModelInfo,
@ -128,6 +129,22 @@ def test_completion_cost_uses_response_model_for_dynamic_routing(_local_model_co
assert cost > 0, "Cost should be calculated using response model"
def test_jina_rerank_bills_total_tokens_at_input_rate_only(_local_model_cost_map):
response: Final = RerankResponse(
id="rerank-1",
results=[{"index": 0, "relevance_score": 0.9}],
meta={"billed_units": {"total_tokens": 1000}},
)
cost: Final = completion_cost(
completion_response=response,
model="jina_ai/jina-reranker-v2-base-multilingual",
call_type="rerank",
)
assert cost == pytest.approx(1000 * 5e-08)
def test_cost_calculator_with_response_cost_in_additional_headers():
class MockResponse(BaseModel):
_hidden_params = {