scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-08-28 05:25:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 36

fix(map): stop claiming max reasoning effort for gpt-5.6 on chat completions

Browse source 
/v1/chat/completions rejects reasoning_effort=max on every gpt-5.6 snapshot with
"Unsupported value: 'reasoning_effort' does not support 'max' with this model.
Supported values are: 'none', 'low', 'medium', 'high', and 'xhigh'", so the 16
gpt-5.6 entries that asserted supports_max_reasoning_effort had model groups
advertising a level routing would always get a 400 for.

The flag reads as chat-surface capability everywhere else in the map: before
this branch no openai or azure entry carried it at all, only anthropic-family
ones whose chat endpoint really does take max. Dropping it lines the gpt-5.6
advertisement up with what the endpoint accepts and lets the chat gate refuse
the level with our own error instead of forwarding it into a provider 400.

/v1/responses does accept max for gpt-5.6 and is unaffected: the responses
config gates none alone, so the cursor thinking-max variant keeps resolving.
This commit is contained in:
mateo-berri 2026-08-21 18:25:39 -07:00 • committed by Tin Chi Lo
parent b0fee551fd
commit 6ec49ea7dd
3 changed files with 57 additions and 69 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

48
litellm/model_prices_and_context_window_backup.json
View file

@ -6639,8 +6639,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/gpt-5.6-sol": {
"cache_read_input_token_cost": 5e-07,
@ -6691,8 +6690,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/gpt-5.6-terra": {
"cache_read_input_token_cost": 2e-07,
@ -6743,8 +6741,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/gpt-5.6-luna": {
"cache_read_input_token_cost": 2e-08,
@ -6795,8 +6792,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/us/gpt-5.6": {
"cache_read_input_token_cost": 5.5e-07,
@ -6843,8 +6839,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/us/gpt-5.6-sol": {
"cache_read_input_token_cost": 5.5e-07,
@ -6892,8 +6887,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/us/gpt-5.6-terra": {
"cache_read_input_token_cost": 2.2e-07,
@ -6941,8 +6935,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/us/gpt-5.6-luna": {
"cache_read_input_token_cost": 2.2e-08,
@ -6990,8 +6983,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/eu/gpt-5.6": {
"cache_read_input_token_cost": 5.5e-07,
@ -7038,8 +7030,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/eu/gpt-5.6-sol": {
"cache_read_input_token_cost": 5.5e-07,
@ -7087,8 +7078,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/eu/gpt-5.6-terra": {
"cache_read_input_token_cost": 2.2e-07,
@ -7136,8 +7126,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/eu/gpt-5.6-luna": {
"cache_read_input_token_cost": 2.2e-08,
@ -7185,8 +7174,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/gpt-5.5": {
"deprecation_date": "2027-10-26",
@ -26417,8 +26405,7 @@
"supports_tool_choice": true,
"supports_vision": true,
"supports_web_search": true,
"supports_xhigh_reasoning_effort": true,
"supports_max_reasoning_effort": true
"supports_xhigh_reasoning_effort": true
},
"gpt-5.6-sol": {
"cache_creation_input_token_cost": 5e-06,
@ -26482,8 +26469,7 @@
"supports_tool_choice": true,
"supports_vision": true,
"supports_web_search": true,
"supports_xhigh_reasoning_effort": true,
"supports_max_reasoning_effort": true
"supports_xhigh_reasoning_effort": true
},
"gpt-5.6-terra": {
"cache_creation_input_token_cost": 2.5e-06,
@ -26546,8 +26532,7 @@
"supports_tool_choice": true,
"supports_vision": true,
"supports_web_search": true,
"supports_xhigh_reasoning_effort": true,
"supports_max_reasoning_effort": true
"supports_xhigh_reasoning_effort": true
},
"gpt-5.6-luna": {
"cache_creation_input_token_cost": 2.5e-07,
@ -26610,8 +26595,7 @@
"supports_tool_choice": true,
"supports_vision": true,
"supports_web_search": true,
"supports_xhigh_reasoning_effort": true,
"supports_max_reasoning_effort": true
"supports_xhigh_reasoning_effort": true
},
"gpt-5.6-cyber": {
"cache_creation_input_token_cost": 1.5625e-05,

48
model_prices_and_context_window.json
View file

@ -6639,8 +6639,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/gpt-5.6-sol": {
"cache_read_input_token_cost": 5e-07,
@ -6691,8 +6690,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/gpt-5.6-terra": {
"cache_read_input_token_cost": 2e-07,
@ -6743,8 +6741,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/gpt-5.6-luna": {
"cache_read_input_token_cost": 2e-08,
@ -6795,8 +6792,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/us/gpt-5.6": {
"cache_read_input_token_cost": 5.5e-07,
@ -6843,8 +6839,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/us/gpt-5.6-sol": {
"cache_read_input_token_cost": 5.5e-07,
@ -6892,8 +6887,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/us/gpt-5.6-terra": {
"cache_read_input_token_cost": 2.2e-07,
@ -6941,8 +6935,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/us/gpt-5.6-luna": {
"cache_read_input_token_cost": 2.2e-08,
@ -6990,8 +6983,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/eu/gpt-5.6": {
"cache_read_input_token_cost": 5.5e-07,
@ -7038,8 +7030,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/eu/gpt-5.6-sol": {
"cache_read_input_token_cost": 5.5e-07,
@ -7087,8 +7078,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/eu/gpt-5.6-terra": {
"cache_read_input_token_cost": 2.2e-07,
@ -7136,8 +7126,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/eu/gpt-5.6-luna": {
"cache_read_input_token_cost": 2.2e-08,
@ -7185,8 +7174,7 @@
"supports_web_search": true,
"supports_none_reasoning_effort": true,
"supports_xhigh_reasoning_effort": true,
"supports_minimal_reasoning_effort": false,
"supports_max_reasoning_effort": true
"supports_minimal_reasoning_effort": false
},
"azure/gpt-5.5": {
"deprecation_date": "2027-10-26",
@ -26417,8 +26405,7 @@
"supports_tool_choice": true,
"supports_vision": true,
"supports_web_search": true,
"supports_xhigh_reasoning_effort": true,
"supports_max_reasoning_effort": true
"supports_xhigh_reasoning_effort": true
},
"gpt-5.6-sol": {
"cache_creation_input_token_cost": 5e-06,
@ -26482,8 +26469,7 @@
"supports_tool_choice": true,
"supports_vision": true,
"supports_web_search": true,
"supports_xhigh_reasoning_effort": true,
"supports_max_reasoning_effort": true
"supports_xhigh_reasoning_effort": true
},
"gpt-5.6-terra": {
"cache_creation_input_token_cost": 2.5e-06,
@ -26546,8 +26532,7 @@
"supports_tool_choice": true,
"supports_vision": true,
"supports_web_search": true,
"supports_xhigh_reasoning_effort": true,
"supports_max_reasoning_effort": true
"supports_xhigh_reasoning_effort": true
},
"gpt-5.6-luna": {
"cache_creation_input_token_cost": 2.5e-07,
@ -26610,8 +26595,7 @@
"supports_tool_choice": true,
"supports_vision": true,
"supports_web_search": true,
"supports_xhigh_reasoning_effort": true,
"supports_max_reasoning_effort": true
"supports_xhigh_reasoning_effort": true
},
"gpt-5.6-cyber": {
"cache_creation_input_token_cost": 1.5625e-05,

30
tests/test_litellm/llms/openai/test_gpt5_transformation.py
View file

@ -1311,14 +1311,34 @@ def test_responses_gpt54_allow_temperature_effort_none(
assert params["temperature"] == 0.7
def test_gpt5_6_allows_reasoning_effort_max(config: OpenAIConfig):
params = config.map_openai_params(
non_default_params={"reasoning_effort": "max"},
optional_params={},
@pytest.mark.parametrize("model", ["gpt-5.6", "gpt-5.6-sol", "gpt-5.6-terra", "gpt-5.6-luna"])
def test_gpt5_6_rejects_reasoning_effort_max_on_chat_completions(config: OpenAIConfig, model: str):
"""/v1/chat/completions answers max with "Unsupported value: 'reasoning_effort' does not support
'max' with this model. Supported values are: 'none', 'low', 'medium', 'high', and 'xhigh'" on
every gpt-5.6 snapshot, so no gpt-5.6 map entry asserts supports_max_reasoning_effort and the
level is refused here instead of being forwarded into a provider 400."""
with pytest.raises(litellm.utils.UnsupportedParamsError):
config.map_openai_params(
non_default_params={"reasoning_effort": "max"},
optional_params={},
model=model,
drop_params=False,
)
def test_gpt5_6_keeps_reasoning_effort_max_on_the_responses_api(
responses_config: OpenAIResponsesAPIConfig,
):
"""/v1/responses accepts max for gpt-5.6, and that is the surface the cursor thinking-max
variant resolves onto, so the responses path keeps carrying the level chat refuses."""
params = responses_config.map_openai_params(
response_api_optional_params=ResponsesAPIOptionalRequestParams(
reasoning={"effort": "max"},
),
model="gpt-5.6",
drop_params=False,
)
assert params["reasoning_effort"] == "max"
assert params["reasoning"] == {"effort": "max"}
def test_gpt5_6_rejects_reasoning_effort_ultra_until_the_map_opts_in(config: OpenAIConfig):