Fix Ollama agent replies after tool calls

This commit is contained in:
russo94 2026-07-28 21:22:18 +02:00
parent 01790c2f08
commit e3eab6734a

View file

@ -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)