Skip to content

Fix tool search failing with reasoning models #44

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
vl3c merged 1 commit into from
Feb 16, 2026
Merged

Fix tool search failing with reasoning models #44

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 7 additions & 1 deletion static/tool_search_service.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -171,9 +171,15 @@ def search_tools(
# Clamp max_results to valid range
max_results = max(1, min(20, max_results))

# Use provided model, instance default, or fallback to gpt-4.1-mini
# Use provided model, instance default, or fallback to gpt-4.1-mini.
# Tool search is a lightweight classification task — reasoning models
# are overkill and their max_completion_tokens budget includes internal
# reasoning tokens, so 500 tokens may not leave enough room for output.
# For OpenAI reasoning models, downgrade to gpt-4.1-mini automatically.
if model is None:
model = self.default_model or AIModel.from_identifier("gpt-4.1-mini")
if getattr(model, "is_reasoning_model", False) and getattr(model, "provider", "") == "openai":
model = AIModel.from_identifier("gpt-4.1-mini")

# Build the prompt
tool_descriptions = self.build_tool_descriptions()
Expand Down