Unverified85% confidenceFactTime unknown
MLC LLM和llama.cpp等端侧推理框架已使在手机本地运行小型语言模型成为可能
1
Sources
85%
Confidence
Long-term
Relevance
6/1/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedSmall language models typically refer to language models with fewer than 1B parameters; representative works include Microsoft's Phi series (Phi-1 to Phi-3, 1.3B–3.8B parameters), Google's Gemma 2B, and MobileLLM68% similarUnverifiedMicrosoft is advancing the deployment of the small language model Phi series on local devices for on-device inference.68% similarUnverified大语言模型的执行层通过 Function Calling（工具调用）机制将 LLM 的文本输出转化为对真实系统的操作指令68% similarUnverifiedLlamaFactory supports unified fine-tuning for more than 100 large language models (LLMs) and vision-language models (VLMs)67% similarUnverified扩散语言模型（dLLM）支持双向注意力和任意顺序生成，不必像GPT那样从左到右逐个输出token67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/19476API
curl https://kongchang.com/api/v1/knowledge/claims/19476MCP
get_claim(id=19476)