From e3eab6734a57f719669a5f2d638a535286d8b689 Mon Sep 17 00:00:00 2001
From: russo94 <216342474+russo94@users.noreply.github.com>
Date: Tue, 28 Jul 2026 21:22:18 +0200
Subject: [PATCH] Fix Ollama agent replies after tool calls

---
 src/llm_core.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/src/llm_core.py b/src/llm_core.py
index 4dec32376..f99e705e5 100644
--- a/src/llm_core.py
+++ b/src/llm_core.py
@@ -2228,6 +2228,8 @@ async def _stream_llm_inner(url: str, model: str, messages: List[Dict], temperat
         # <think> blocks. Ollama /v1 accepts "think": false as a top-level param.
         if _is_ollama_openai_compat_url(url) and _supports_thinking(model):
             payload["think"] = False
+            if tools:
+                payload["reasoning_effort"] = "none"
         _apply_local_cache_affinity(payload, url, session_id)
         _apply_local_generation_stability(payload, target_url, model)
         _scrub_openai_chat_tool_reasoning(payload, target_url, model)