scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-09-09 22:31:41 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 86

fix(cost-map): stop advertising reasoning_effort max on azure_ai/gpt-6-astra

Browse source 
Foundry rejects reasoning_effort max on the gpt-6-astra deployment with a 400 that
names none, low, medium, high, and xhigh as the supported values, so the card no
longer lists max. The request path never gated max (only xhigh is opt-in), so this
only changes /model_group/info and router capability gating. The azure/ twin stays
as is because it was not verified on an Azure OpenAI host
This commit is contained in:
mateo-berri 2026-09-05 19:42:15 -07:00
parent a17fcecf70
commit e8f311429e
3 changed files with 15 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
litellm/model_prices_and_context_window_backup.json
View file

@ -3518,7 +3518,7 @@
],
"supports_computer_use": true,
"supports_function_calling": true,
"supports_max_reasoning_effort": true,
"supports_max_reasoning_effort": false,
"supports_minimal_reasoning_effort": false,
"supports_native_streaming": true,
"supports_none_reasoning_effort": true,

2
model_prices_and_context_window.json
View file

@ -3518,7 +3518,7 @@
],
"supports_computer_use": true,
"supports_function_calling": true,
"supports_max_reasoning_effort": true,
"supports_max_reasoning_effort": false,
"supports_minimal_reasoning_effort": false,
"supports_native_streaming": true,
"supports_none_reasoning_effort": true,

14
tests/test_litellm/router_utils/test_reasoning_effort_capability.py
View file

@ -394,7 +394,6 @@ class TestGpt6AstraAdvertisesItsDocumentedLevels:
[
("azure/gpt-6-astra", "azure"),
("azure/us/gpt-6-astra", "azure"),
("azure_ai/gpt-6-astra", "azure_ai"),
],
)
def test_a_foundry_deployment_also_advertises_none(self, local_model_cost_map, model, custom_llm_provider):
@ -413,3 +412,16 @@ class TestGpt6AstraAdvertisesItsDocumentedLevels:
"xhigh",
"max",
)
def test_a_foundry_azure_ai_deployment_advertises_none_but_not_max(self, local_model_cost_map):
from litellm.utils import _get_model_info_helper
model_info = dict(_get_model_info_helper(model="azure_ai/gpt-6-astra", custom_llm_provider="azure_ai"))
assert resolve_supported_reasoning_efforts(model_info, deployment_is_mapped=True) == (
"none",
"low",
"medium",
"high",
"xhigh",
)