Expired50% confidenceFactExact time
The new Siri adopts a model routing strategy, where simple local tasks are handled by an on-device small model of approximately 3 billion parameters, while complex tasks are routed to Google Gemini or OpenAI ChatGPT cloud APIs.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/15/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 10/13/2026(expired)
Sources
Related Claims
UnverifiedSkillSpector's model-assisted mode supports connecting to cloud-based large language models as well as locally hosted services compatible with the OpenAI interface.73% similarUnverifiedOpenAI's Assistants API, Anthropic's tool-calling interface, and Google Gemini's Agent capabilities continue to iterate.69% similarUnverifiedOpenAI、Google、Anthropic等基础模型公司通过推出ChatGPT Plugins、Gemini Extensions等功能向下游延伸，侵蚀套壳应用的生存空间68% similarUnverifiedThe training frameworks for models such as OpenAI's GPT series, Meta's LLaMA, and Google's Gemini are all based on PyTorch or JAX (Python interface).68% similarUnverifiedOpenScaler achieves token cost control through local model routing and MetaSkill task orchestration, and is positioned closer to the infrastructure layer rather than a complete resident Agent OS.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/519174API
curl https://kongchang.com/api/v1/knowledge/claims/519174MCP
get_claim(id=519174)