klbr/khoj
1
0
Fork 0
mirror of https://github.com/khoaliber/khoj.git synced 2026-03-02 21:19:12 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix openai reasoning model handling

Browse Source 
- Fix identifying gpt-oss as openai reasoning model
- Drop unsupported stop param for openai reasoning models
- Drop the Formatting re-enabled logic for openai reasoing only models
  We use responses api for openai models and latest openai models are
  hybrid models, they don't seem to need this convoluted system
  message to format response as markdown
This commit is contained in:
Debanjum
2025-08-28 00:49:23 -07:00
parent be79b8a633
commit 9395c17f34
1 changed files with 3 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
src/khoj/processor/conversation/openai/utils.py
View File

@@ -118,7 +118,9 @@ def completion_with_backoff(
model_kwargs["temperature"] = 1
reasoning_effort = "medium" if deepthought else "low"
model_kwargs["reasoning_effort"] = reasoning_effort
# Remove unsupported params for reasoning models
model_kwargs.pop("top_p", None)
model_kwargs.pop("stop", None)
elif is_twitter_reasoning_model(model_name, api_base_url):
model_kwargs.pop("temperature", None)
reasoning_effort = "high" if deepthought else "low"
@@ -304,19 +306,6 @@ async def chat_completion_with_backoff(
# Remove unsupported params for reasoning models
model_kwargs.pop("top_p", None)
model_kwargs.pop("stop", None)
# Get the first system message and add the string `Formatting re-enabled` to it.
# See https://platform.openai.com/docs/guides/reasoning-best-practices
if len(formatted_messages) > 0:
system_messages = [
(i, message) for i, message in enumerate(formatted_messages) if message["role"] == "system"
]
if len(system_messages) > 0:
first_system_message_index, first_system_message = system_messages[0]
first_system_message_content = first_system_message["content"]
formatted_messages[first_system_message_index]["content"] = (
f"{first_system_message_content}\nFormatting re-enabled"
)
elif is_twitter_reasoning_model(model_name, api_base_url):
reasoning_effort = "high" if deepthought else "low"
# Grok-4 models do not support reasoning_effort parameter
@@ -851,7 +840,7 @@ def is_openai_reasoning_model(model_name: str, api_base_url: str = None) -> bool
return (
is_openai_api(api_base_url)
and (model_name.lower().startswith("o") or model_name.lower().startswith("gpt-5"))
or model_name.lower().startswith("gpt-oss")
or "gpt-oss" in model_name.lower()
)