Unverified50% confidenceFactExact time
主流边缘推理框架包括谷歌的TensorFlow Lite、微软主导的ONNX Runtime、以及陈天奇团队开发的MLC-LLM
1
Sources
50%
Confidence
Long-term
Relevance
9/1/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedGoogle开发了TensorFlow Lite和LiteRT等轻量级推理框架用于移动设备端AI推理78% similarUnverified苹果的 Core ML 框架和 Google 的 TensorFlow Lite 为端侧部署提供优化工具链76% similarUnverifiedMeta的GPTQ算法和谷歌的AWQ算法已被集成到llama.cpp、TensorFlow Lite等主流端侧推理框架中72% similarUnverifiedLiteRT-LM 是 Google 面向端侧语言模型推理的运行时，LiteRT 前身为 TensorFlow Lite，Google 于 2024 年将其更名并扩展适用范围71% similarUnverifiedGoogle为Gemma提供了多种量化版本，并通过MediaPipe和TensorFlow Lite等框架支持在Android、iOS、浏览器等环境中运行71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/835581API
curl https://kongchang.com/api/v1/knowledge/claims/835581MCP
get_claim(id=835581)